UNIVERSITÀ DEGLI STUDI DI NAPOLI

“FEDERICO II”





International PhD program on Novel Technologies for
Materials, Sensors and Imaging
XXII cycle



Monitoring of laser – biological tissue
interaction on plane angioma by
functional infrared imaging


Massimo Rippa






Supervisor:
Dr. Pasqual e Mormile





___________________________________________

PhD THESIS – December 2009






For Veronica,
the light of my life
TABLE OF CONTENTS




Introduction......................……………………........................................................…............. i-iv


Chapter 1

LASER – BIOLOGICAL TISSUE INTERACTION


1.1 Introduction……........................................................................................ 1
1.2 Reflection and refraction of the tissue....................................................... 2
1.3 Absorption of the biological tissue............................................................ 3
1.4 Scattering................................................................................................... 5
1.5 Turbid media............................................................................................................ 7
1.6 Laser-Tissue: Interaction Mechanisms...................................................... 9
1.6.1 Thermal Interaction......................................................................... 10
1.6.2 Photochemical Interaction............................................................... 19
1.6.3 Photoablation……………………………………………………... 20
1.6.4 Plasma-induced Ablation……………………………………….... 23
1.6.5 Photodisruption…………………………………………………... 25
1.7 Laser Applications in the medical field…………………………………. 27



Chapter 2

INFRARED IMAGING: Principles and applications


2.1 Introduction………... ……………………………………………………. 32
2.2 Passive thermography…………………………………………………..... 34
2.3 Active thermography: Testing procedures…………………………......... 35
2.3.1 Pulsed Thermography…………………………………………..... 37
2.3.1.1 The complete thermogram sequence…………………… 40
2.3.1.2 Definitions of thermal contrast………………………..... 42
2.3.2 Step Heating (Long Pulse)……………………………………...... 47
2.3.3 Lock-in Thermography…………………………………………… 49
2.3.4 Vibrothermography…………………………………………….… 50
2.4 Infrared detectors………………………………………………….……... 52
2.4.1 Atmospheric Trasmittance……………………………………….. 53
2.4.2 Bolometers……………………………………………………...… 55
2.4.3 Pyroelectric detectors…………………………………………..… 56
2.4.4 Quantum Detectors……………………………………………….. 57
2.5 Infrared Imaging Devices: Focal Plane Arrays (FPA)…………………… 58
2.5.1 Detectors: Performance parameters………………………………. 60
2.6 New approach of infrared imaging in medical science…………………. 62




Chapter 3

INFRARED MONITORING ON PLANE ANGIOMA: PT model and numerical simulation


3.1 Introduction……………………………………………………………..…. 65
3.2 Infrared monitoring of plane angioma treatment: approach description….. 65
3.3 Plane angioma pathology (Port Wine Stain)…………………………........ 70
3.4 Pulsed Thermography model……………………………………………… 72
3.4.1 Formulation of the forward problem: 3D thermal quadrupoles......... 73
3.4.2 Resolution of the forward problem: mathematical perturbations….. 77
3.4.3 Evaluation of the depth…………………………………………….. 80
3.4.4 Evaluation of the thickness................................................................ 81
3.4.5 Statistical analysis of the accuracy of the inversion procedures…… 82
3.5 Numerical simulation: biometric heat transfer mod el................................... 84
3.5.1 Model Computation…………………............................................... 86




Chapter 4

EXPERIMENTAL AND NUMERICAL RESULTS

4.1 Introduction……………………………………………….………….…..... 88
4.2 Materials and method….……………………………………….…….…..... 89
4.3 Preliminary results and observations..……………………………….......... 92
4.4 Pulsed Thermography model: experimental results ..……….….…………. 97
4.5 Numerical simulations: results……………………..…………….….…….. 100
4.6 Limit conditions for applications of laser treatments……..……….………. 116
4.7 Final conclusions: set of laser parameters…...…………………….………. 117






General Conclusions....................................…………………....….......……………… 120



References.....................................................…….……………....….......……………… 125









i
Introduction



The application field that attracted a great deal of attention has been the medicine since the
birth of the laser. From the beginning it was considered as a very potential and promising tool
for several different application in medicine. Nevertheless nobody could have imagined at that
time the development and the enormous effect of its use in so many different sectors of
medicine ranging from surgery to oncology, from urology to ophtalmology, from ginecology to
dermatology, from physiotherapy to aesthetic medicine. Nowadays the use of laser has become
a very common practice in all the medicine sectors and with different pathologies. These
pathologies have been cured thanks to the use of different laser sources, selected as a function
of both wavelengths and power, and to more and more sophisticated technologies. We can just
think to very complicated surgery operation which are performed in a successful way with the
help of laser-scalpel, or to skin tumours treated with coherent light, to operation with laser
endoscopy which is now the common practice, to get up to aesthetic, ophtalmology, and
dermatological operations, and so on.

Dermatology is one of the medical field in which laser action has been succesfully employed.
Deseases like skin melanoms and angioma are usually treated with laser theraphy. In this field,
during the last decades, technology supplied laser sources with technical properties specifically
designed in order to accomplish the specific requirements of several application. Nowadays
dermatologists can rely on a wide range of lasers which vary in wavelengths (some of them out
of the visible range) and thus they can be selectively employed according to the specific
application and type of pathology. The activity of the doctor consists in applying the laser light
for a specific time and with a fixed intensity in the area of interest of the skin of the patient.
After, he verifies the effects induced by the light laser after one or two weeks time. The
physician usually makes use of the laser on his patient in an empirical way. Basically, just
according to his own experience he decides the duration of the application and the intensity of
laser light on the skin according to some parameters such as the individual patient, the
pathology and the skin phototype. In most of the treatments what happens is that the doctor has
no possibility to directly evaluate the effect of the laser applied on line since the wavelength of
ii
it is not in the visible range or it is pulsed. This means flashes of few milliseconds too fast to be
recorded by the eyes. This limits the operation of the doctor because the beam power chosen
might be too low or too high on one side or the duration of the pulse too short or too long
respectively. The information comes only after a few days or weeks later to the physician. And
this seems to be the only remarkable limit in the application of laser in dermatology. It is
important to underline that the doctor should always work under a safe regime for the patient.
This means that he has to avoid damages like scars which might occur since he has no direct
vision of the interaction light-treated tissues and so he usually prefers working with a lower
intensity of light. This aspect, on one side means a safer situation for the patient but on the
other side minimize the efficacy of the therapy making the cure time longer. The possibility to
give to the doctor a tool which makes him able to monitor on-line the reaction of the skin
during the laser application (and not just examining the post application effects) means:

1- To enhance the performances of the light effect enlarging remarkably the application fields
2- To optimize the action in itself in terms of both efficiency and curing times.

In photothermal regime, since the effect of the application of laser light on skin is linked to the
local temperature induced, it is possible to evaluate the effects of such an interaction studying
the thermal dynamic and in general the thermal behaviour of the tissues.
In this view infrared thermography seems to be the best solution. Thanks to this technique, it is
possible to achieve information about thermal interaction of surface and subsurface structure of
the target under investigation by analysis and images elaboration.
In the laser therapy field, infrared imaging technology can be used for a real-time monitoring
of the laser-biological tissue interaction. This technique makes possible the direct evaluation
about the reaction of the tissue in the specific region treated.
The equipment capable of performing such measures is the infrared camera since it supplies
thermal images which represent the dynamic thermal evolution laser induced.

In the research activity of this work-thesis, the main pathology monitored and studied by
infrared technique has been the Plane Angioma known as Port Wine Stain (PWS) pathology.
This pathology is a vascular malformation that consists of a blood vessels accumulation under
the tissue. This accumulation form a subsurface plane of vessels of depth and thickness
depending on the specific anatomic region and, in general, different from case to case.
iii
It’s possible to obtain information about the specific plane of blood vessels under treatment
monitoring the thermal reaction of the tissue under laser action by IR imaging technique.
The main target of this study are:

- To evaluate the support of the infrared imaging, in the dermtological field, as a tool for
scientific investigation

And in particular way

- To evaluate the employment of this technique as a tool for the control and optimization of
laser parameters in order to make the treatment safer and more efficient

In this work a novel methodological approach for the laser parameters optimization in the
treatment on plane angioma is proposed. In particular, with this approach, an active
thermography model, present in the literature, is employed in order to find morphological
information for a specific plane angioma.

With this model from the temporal evolution of the temperature obtained on the tissue after
a single laser pulse-test (heat souce), it is possible to get a local evaluation of the depth and
thickness of the plane of blood vessels in the anatomic area that the physician wants to treat.
This information permits a morphological reconstruction of the multilayer biological tissue
under treatment. Subsequently, a 2D+1 numerical simulation based on a biometric heat
transfer model is executed on this multilayer. With the simulations the thermal behaviour of
the biological multilayers can be achieved for different set of laser parameters as input.

By the evaluations obtained with this approach, the operator can adjust the laser parameters
for each specific therapy in a proper way changing the intensity and the duration of the laser
ligth according to the type of plane of vessels and to the skin phototype treated. This allows
to specialists to employ laser systems in a more specific, well controlled and less empiric way.

The research activity, theoretical and experimental, object of this thesis work, is described in
four chapters.

iv
In the first chapter the foundamental theoretical concept of the laser–tissue interaction are
presented. In particular, the main interaction mechanisms as the photothermal interaction, the
photochemical interaction, the photoablation, the plasma-induced ablation and the
photodisruption are described. The chapter conclude with the state of art of the laser
application in medical field.

In the second chapter the priciples and applications of the infrared imaging technique are
reported. The passive and active approach are explained. The problems relative to atmospheric
trasmittance and infrared detectors are treated. At the end of the chapter, our novel approach
and , in particular, the new application of infrared imaging in medical science are discussed.

The third chapter, is the reading key for the understanding of the experimental work presented in
this thesis. A detailed description of the activity research is given. The novel approach proposed
and realized, the pathology analized and computation instruments used are treated and
discussed. Particularly, the pulsed thermography model and the numerical approach employed
in the operative protocol performed are explained in detail.

In the last chapter, the main experimental and numerical results found are reported and
discussed. The main characteristic of the laser tool and the infrared cameras used in the
monitoring are showed. Some preliminaries results are presented. Numerical and experimental
results achieved with the novel approach proposed and relative to different multilayer
biological tissues affected by plane angioma are illustrated and compared. In the conclusions of
the chapter, some set of laser parameters optimized for treatment of specific clinical situations
are proposed.

The experimental research activity has been made in collaboration with the Department of
Dermatology of the University II Policlinico in Naples, monitoring patients in the laser therpay
and photodynamics ambulatory. The data analysis and imaging elaboration have been done in
the LIRT of the Cibernetic Institute of CNR “E. Caianiello” in Pozzuoli (Na), which funded
my grant, under the supervision of Dr. P. Mormile and in collaboration with Dr. L. Petti to
whom go my most sincere thanks.
In addition I also want to thank Dr. A. Baldo and Dr. G. Monfregola of the Department of
Dermatology of the University II Policlinico for their availability, support and courtesy.
1
Chapter 1

LASER – BIOLOGICAL TISSUE INTERACTION

1.1 Introduction

In this chapter, it will be discussed basic phenomena occurring when matter is exposed to light.
Matter can act on electromagnetic radiation in manifold ways. When a light beam is incident
on a slice of matter, in principle, three effects exist which may modify its propagation [1-4]:

- reflection and refraction
- absorption
- scattering

In medical laser applications, however, refraction plays a significant role only when irradiating
transparent media (corneal tissue). In opaque media, it is, usually, difficult to measure the
effect of refraction due to absorption and scattering. According to the type of material and the
incident wavelenght, we can have a predominance of either losses or reflection, absorption or
scattering. The optical parameters as the refractive index, the absorption and scattering
coefficient depend on the wavelength of the incident light. In particular, the index of refractive
strongly depends on wavelength in spectral regions of high absorption for the tissue only.









Fig 1.1 : Geometry of reflection, refraction, absorption and diffused radiation
Reflection
Incident radiation
Absorption and
diffused radiation
2

1.2 Reflection and refraction of the tissue

In laser surgery, knowledge of absorbing and scattering properties of a selected tissue is
essential in order of predicting a successful treatment. In reality, of course, limitations are
given by the inhomogeneity of biological tissue which are also responsible for our inability to
provide more than mean parameters of the tissues.

Reflection is defined as the returning of electromagnetic radiation by surfaces upon which it is
incident. In general, a reflecting surface is the physical boundary between two materials of
different indices of refraction such as air and tissue. The simple law of reflection states that the
reflection angle ` equals the angle of incidence , expressed by:


'
u u = 1.1

In contrast, when the roughness of the reflecting surface is comparable or even larger than the
wavelenght of radiation, diffuse reflection occurs. Diffuse reflection is a common phenomenon
of all tissues , since none of them is a highly polished surface such as an optical mirror.
Refraction occurs when the reflecting surface separates two media of different indices of
refraction (i.e. n
1
and n
2
). The mathematical relation governing refraction is known as Snell’s
law :

2 2 1 1
sin sin u u n n = 1.2

When
1 2 1
/ sin n n > u the relation (1.2) is not fulfilled and a total reflection occurs.


Fig. 1.2: Geometry of light propagation
3
1.3 Absorption of the biological tissue

Absorption is due to a partial conversion of light energy into heat motion of vibrations of
molecules of the absorbing material. A perfectly transparent medium permits the passage of
light without any absorption. Among biological tissue cornea and lens can be considered as
being highly transparent for visible light. In contrast, media in which incident radiation is
reduced prectically to zero are called opaque.
The terms “transparent” and “opaque” are relative, since they are wavelength dependent.
Actually, no tissue is known to be either transparent or opaque to all wavelenghts of the
electromagnetic spectrum.
The law that descibes the effect of thickness on absorption is called Lambert-Beer law and is
expressed by:

) exp( ) (
0
z I z I o ÷ = 1.3

where z denotes the optical axis, I(z) is the intensity at a distance z, I
0
is the incident intensity,
is the absorption coefficient of the medium. The inverse of the absorption coefficient is
also referred to as the absorption length L

L=1/ 1.4

The absorption length measures the distance z in which the intensity I(z) has dropped to 1/e of
its incident value I
0
.
In biological tissues, absorption is mainly caused by either water molecules or macromolecules
such as proteins and pigments. Whereas absorption in the IR region of the spectrum can be
primarily attributed to water molecules, proteins as well as pigments mainly absorb in the UV
and visible range of the spectrum.
In fig 1.3, absorption spectra of two elementary biological absorbers are shown. They belong to
melanin and hemoglobin (HbO
2
), respectively.
Melanin is the basic pigment of skin and is by far the most important epidermal chromophore.
Its absorption coefficient monotonically increases across the visible spectrum toward the UV.
Haemoglobin is predominant in vascularized tissues. It has relative absorption peaks around
4
280 nm, 420 nm, 540 nm and 580 nm, and then exhibits a cut-off at approximately 600 nm.
Since the green and yellow wavelengths of Argon and Krypton ion laser perfectly match the
absorption peaks of haemoglobin, these lasers can be used for the coagulation of blood and
blood vessels. For some clinical applications, dye lasers may be an alternative choice since
their tunability can be advantageously used to match particular absorption bands of specific
proteins and pigments (selective absorption) .
Since neither macromolecules nor water strongly absorb in the near IR , a therapeutic window
for many patologies is delineated between roughly 600 nm and 1200 nm. In this range,
radiation penetrates biological tissues at lower loss, thus enabling treatment of deeper tissue
structures.



Fig. 1.3: Absorption spectra of melanin in skin and haemoglobin (HbO
2
) in blood










5
1.4 Scattering

Scattering is a class of phenomena by which particles are deflected by collisions with other
particles. In paricular, it takes place when charged particles are exposed to electromagnetic
waves of frequencies not corresponding to those natural frequency of free vibration of the same
particles (resonance frequencies). Scattering can be elastic or inelastic, depending on whether
part of the incident photon energy is converted during the process of scattering. Elastic
scattering occurs when incident and scattered photons have the same energy. A special kind of
elastic scattering is Rayleigh scattering. Its only restriction is that the scattering particles are
smaller than the wavelenght of the incident radiation. In particular, it’s possible to find a
relationship between scattered intensity and refraction index. This statement is also known as
Rayleigh’s law and is given by:


4
2
) 1 (
ì
÷
~
n
I
S
1.5

If the scattering angle is taken into account, a more detailed analysis yelds


4
) cos( 1
ì
u +
~
S
I 1.6

where =0 denotes forward scattering. Rayleigh’s law is illustrated in fig. 1.4


Fig. 1.4: Rayleigh’s law of scattering in three different regions
6
Within the visible spectrum, scattering is significantly reduced when we shift from blu to red
light. In the derivation of Rayleigh’law, (1.5) and (1.6), absorption has been neglected.
Therefore the relations given are valid only for wavelengths far away from any absorption band.
If the size of the scattering particles becomes comparable to the wavelenght of the incident
radiation, such as in the blood cells, Rayleigh scattering no longer applies and another type of
scattering called Mie scattering occurs. The theory of Mie scattering is rather complex and will
not be repeated here. However, it is well known that Mie scattering and Rayleigh scattering
differ in two important aspects. First, Mie scattering shows a weaker dependeance on
wavelenght (~
-x
with 0.4 x 0.5) compared to Rayleigh scattering (~
-4
). Second, Mie
scattering preferably takes place in the forward direction, whereas Rayleigh scattering is
proportional to 1+cos
2
( ) according to (1.6), i.e. forward and backward scattered intensities are
the same.
In most biological tissues, it was found by Wilson and Adam (1983), Jacques et al. (1987), and
Parsa et al. (1989) that photons are preferably scattered in the forward direction. This
phenomenon cannot be explained by Rayleigh scattering. On the other hand, the observed
wavelength dependence is somewhat stronger than that predicted by Mie scattering. Thus,
neither Rayleigh scattering nor Mie scattering completely describe scattering in tissues.
Therefore, it is very convenient to define a probability function p( ) of a photon to be scattered
by an angle which can be fitted to the experimental data. When p( ) does not depend on ,
the scattering is isotropic. Otherwise, anisotropic scattering occurs.
A measure of the anisotropy of scattering is given by the coefficient of anisotropy g, where g =
1 denotes purely forward scattering, g = -1 purely backward scattering, and g = 0 isotropic
scattering. In polar coordinates, the coefficient g is defined by


í
í
=
t
t
e u
e u u
4
4
) (
) cos( ) (
d p
d p
g 1.7

where p( ) is a probability function and o u u e d d d ) sin( = is the elementary solid angle. By
definition , the coefficient of anisotropy g represents the avrerage value of the cosine of the
scattering angle . As a good approximation, it can be assumed that g ranges from 0.7 to 0.99
for most biological tissues. Hence the corresponding scattering angles are most frequently
between 8° and 45°.
7
The important term in 1.7 is the probability function p( ), also called phase function. Several
theoretical phase function p( ) have been proposed in literature in order to find the best
accordance with experimental observation. More detailied information on these phase
functions are provided in the references [5-8].



1.5 Turbid media

In the previous two sections, we have considered the occurrences of either absorption or
scattering. In most tissue both of them are present simultaneously. Such media are called turbid
media. Their total attenuation coefficient can be expressed by


s t
o o o + = 1.8

In turbid media, the mean free optical path of incident photons is thus determined by


s t
t
L
o o o +
= =
1 1
1.9

Only in some cases either or
s
may be negligible with respect to each other, but it is
important to realize the existence of both processes and the fact that usually both are operating.
It is very convenient to define an additional parameter, the optical albedo a , by


t
s
a
o
o
= 1.10

For 0 = a , attenuation is exclusively due to absorption, whereas in the case of 1 = a only
scattering occurs. For 5 . 0 = a , the relation (1.10) can be turned into the equality =
s.
In
general, both effects will take place but they will occur in variable ratios.
8
In fig. 1.4 it is shown the albedo as a function of the scattering coefficient. Three different
absorption coefficient are assumed which are typical of biological tissue. For
s
>>
,
the
albedo asymptotically approaches unity.


Fig. 1.5: Optical albedo as a function of scattering coefficient


In literature, reduced scattering and attenuation coefficient are often used and are given by

) 1 ( ' g
s s
÷ =o o 1.11

and

' '
s t
o o o + = 1.12


since forward scattering alone, i.e. g =1, would not lead to an attenuation of intensity.






9
1.6 Laser-Tissue: Interaction Mechanisms

The interaction mechanisms that may occur when applying laser light to biological tissue are
manifold. Specific laser characteristics as well as laser parameters contribute to this diversity.
Most important among optical tissue properties are the coefficients of reflection, absorption,
scattering and others such as heat conduction and heat capacity. All together, they determine
the total trasmission of the tissue at certain wavelengths. On the other hand, the following
parameters are given by the laser radiation itself: wavelenght, exposure time, energy density
and spot size. Among these, the exposure time is a very crucial parameter when selecting a
certain type of interaction.
Although the number of possible combinations for the experimental parmeters is unlimited,
five main categories of interaction types are classified today : photochemical interaction,
thermal interaction, photoablation, plasma-induced ablation and photodisruption. All these
different interaction types share a single common issue: the characteristic energy density
ranges from approximately 1 to 1000 J/cm
2
. On the other hand, the parameter that ditinguishes
and controls these processes is the exposure time. A map with the five basic interaction types is
shown in fig. 1.6


Fig. 1.6: Map of laser tissue interaction

The y axis expresses the applied power density in W/cm
2
and the exposure time in seconds is
reported on the x axis. According to this map, the time scale can be roughly divided into five
10
sections: continuos wave or exposure times more than 1 s for photochemical interaction, from
1 s to 1 μs for thermal interaction, from 1 μs 1 ns for photoablation, and less than 1 ns for
plasma-induced ablation and photodisruption. The difference between the latter two is
attributed to different energy densities.
The reciprocal correlation between power density and exposure time clearly demonstrates that
roughly the same energy density is required for any type of interaction. Thus, the exposure
time appears to be the main parameter responsible for the variety of interaction mechanisms.
Anyway, adjacent interaction types present in the map cannot always be strictly separated. For
instance, thermal effects may also play an important role during photochemical interaction.
In the next sections each of these interactions mechanisms will be briefly discussed whereas a
greatest attention will be given to the description of the thermal interaction mechanism which
is core of the present thesis work.


1.6.1 Thermal Interaction

The thermal interaction term stands for a large group of interaction types, where the increase in
local temperature induces the significant parameter change. Thermal effects can be induced by
either CW or pulsed laser radiation. Depending on the duration and peak value of the tissue
temperature achieved, different effects like coagulation, vaporization, carbonization, and
melting may be distinguished :

- Coagulation: during the process of coagulation, temperature reaches at least 60°C, and
coagulation tissue becomes necrotic.
- Vaporization: It is a thermomechanical effect due to the pressure build-up involved. This
effect is attributed to the rich presence of water molecules in the tissues. Water, in fact,
strongly absorbs to different wavelength in the NIR region (i.e. 2.94 μm). If a laser with one of
these wavelengths is used (i.e. Er:YAG) a higher pressure in the tissue is induced (water tries
to expand in volume as it vaporizes) and this produce localized microexplosions. The resulting
ablation is called thermal decomposition
- Carbonization: At temperature over 100°C, the tissue starts to carbonize. For medical laser
applications, carbonization should be avoided in any case, since tissue already becomes
necrotic at lower temperatures. Thus, carbonization only reduces visibility during surgery.
11
- Melting: They originate from thermal stress induced by a local temperature gradient across
the tissue surface.
Temperature, obviously, is the governing parameter of all thermal laser-tissue interactions. At
the microscopic level, thermal effect have their origin in bulk absorption occuring in molecular
vibration-rotation bands followed by nonradiative decay. The reaction can be considered as a
two step process. First, absorption of a photon with an energy h promotes the molecule to an
excited state A*, and second, inelastic collision with some partner of the surrounding medium
lead to a deactivation of A* and a simultaneous increase in the kinetic energy M. Therefore, the
temperature rise microscopically originates from the transfer of photon energy to kinetic
energy. This two step process can be written as

- absorption A+ h  A*
- deactivation A* + M(E
kin
)  A + M(E
kin
+ E
kin
)

Both these steps are highly efficient provided that the duration of laser exposure is properly
selected. In biological tissues, absorption is mainly due to the presence of free water molecules,
proteins, pigments, and other macromolecules and it is governed by Lambert-Beer law. The
absorption coefficient strongly depends on the wavelenght of the incident laser radiation. In
thermal interactions, absorption by water molecules plays a significant role. Therefore, the
absorption spectrum of water, an important constituent of most tissues, is shown in the next
figure


Fig. 1.7: Absorption of water

12
As it is possible to see the absorption of water is extremely small in the visible range while in
the IR range water molecules are the dominant absorbers. In this last region absorption
originates from the presence of resonance frequencies associated to symmetric and asymmetric
vibrational modes of water molecules. Typical absorption coefficients and the corresponding
absorption length L for the most important laser wavelenghts are summarized in table 1.1. The
great attenuation in the UV region is stronghly enhanced by Rayleigh scattering.

















Tab 1.1: Absorption coefficients and absorption length L of water at different wavelength









Wavelength (nm) Laser type (cm
-1
) L (cm)
193 ArF 0.1 10
248 KrF 0.018 55
308 XeCl 0.0058 170
351 XeF 0.0023 430
514 Argon ion 0.00029 3400
633 He-Ne 0.0029 340
694 Ruby 0.0056 180
800 Diode 0.020 50
1053 Nd:YLF 0.57 1.7
1064 Nd:YAG 0.61 1.6
2120 Ho:YAG 36 0.028
2940 Er:YAG 12000 0.00008
10600 CO
2
860 0.001
13
- Heat Generation

Heat generation is determined by laser parameters and optical tissue properties. We assume
that a slab of tissue in air is illuminated by Gaussian-shaped laser beam as illustrated in fig. 1.8.



Fig. 1.8: Geometry of the tissue irradiation


For the sake of simplicity, a cylindrical geometry is chosen with z denoting the optical axis,
and r the distance from this axis. Then, the amplitude of the electric field and the
corresponding intensity inside the tissue are given by


)
8
exp( )
2
exp( ) , , (
)
4
exp( )
2
exp( ) , , (
2
2
2
2
0
2
2
2
2
0
t
o
e
t
o
e
t
z
r
I t z r I
t z r
E t z r E
÷ ÷ ÷ =
÷ ÷ ÷ =
1.13

where E
0
and I
0
are the incident values of electric field and intensity, respectively, is the
beam waist, is the absorption coefficient, and is the pulse duration. The incident values E
0
and I
0
are related to each other by the basic electrodynamic equation


2
0 0 0
2
1
cE I c = 1.14

where
0
c

is the dielectic costant , and c is the speed of light.
By means of the two-step process as described above, heat is generated inside the tissue during
the laser exposure. Deposition of heat in tissue is due only to light that is absorbed in the tissue.
14
For a light flux in the z-direction in a nonscattering medium, the local heat deposition per unit
area and time in a thickness z is given by


z
t z z r I t z r I
t z r S
A
A + ÷
=
) , , ( ) , , (
) , , ( 1.15

And, as z approaches zero, we have that


z
t z r I
t z r S
c
c
÷ =
) , , (
) , , ( 1.16

Therefore, under all circumstances, heat deposition is determined by


) , , ( ) , , ( t z r I t z r S o = 1.17

Thus, the last heat source S(r,z,t) inside the exposed tissue is a function of the absorption
coefficient and the local Intensity I(r,z,t). Since is stronghly wavelength-dependent, the
same applies for S. If phase transitions or tissue alterations do not occur, an alteration in heat
content dQ induces a linear change in temperature dT according to the basic law of
thermodynamics

mcdT dQ = 1.18


where m is the mass, and c is the specific heat capacity. For most tissues a good
approximation is given by

) 8 . 2 55 . 1 (
m
m
w
c + = 1.19

where m is the tissue density and
w
m

is its water content expressed in Kg/m
3


15
- Heat Effects

Approximate values of achievable temperatures can often estimated. The thermal interaction
dealing with biological effects are related to different temperatures inside the tissue. As already
stated, these can be manifold, depending on the type of tissue alterations.
Assuming a body temperature of 37°C, no measurable effects are observed for the next 5°C
above this. The first mechanism by which tissue is thermally affected can be attributed to
conformational changes of molecules. These effects, accompanied by bond destruction and
membrane alterations, are summarized in the single term hyperthermia ranging from
approximately 42-50 °C. If such a hyperthermia lasts for several minutes, a significant
percentage of the tissue will undergo necrosis as described below by Arrhenius’ equation.
Beyond 50°C, a measurable reduction in enzyme activity is observed , resulting in reduced
energy transfer within the cell and immobility of the cell. Furthermore, certain repair
mechanism of the cell are disabled. Thereby, the fraction of surviving cells is further reduced.
At 60°C, denaturation of proteins and collagen occurs which leads to coagulation of tissue and
necrosis of cells. The corresponding macroscopic response is the visible paling of the tissue. At
even higher temperatures more then 80 °C the membrane permeability is drastically increased,
thereby destroying the otherwise maintained equilibrium of chemical concentrations.
At 100°C, water molecules contained in most tissues start to vaporize. The large vaporization
heat of water (2253 KJ/Kg) is advantageous, since the vapor generated carries away excess
heat and helps to prevent any further increase in the temperature of adjacent tissues. Due to the
large increase in volume during this phase transition, gas bubbles are formed inducing
mechanical ruptures and thermal decomposition of tissue fragments. Only if all water
molecules have been vaporized, and laser exposure is still continuing, does the increase in
temperature proceeds. At temperatures exceeding 100°C, carbonization takes place which is
observable by the black-ening of adjacent tissues and the escape of smoke. To avoid
carbonization, the tissue is usually cooled down with either water or gas. Finally, beyond
300°C, melting can occur, depending on the target material.
All these steps are summarized in table 1.2, where the local temperature and the associated
tissue effects are listed



16

Temperature (°C)

Biological effect

43 - 45
- Conformational alteration
- Shrinkage of the tissue
- Hyperthermia

45 - 60
- Alteration of enzymatic activity
- Cell immobility

60 - 75
- Protein denaturation
- Coagulation

75 - 100
- Collagen denaturation
- Permeabilization of membranes

100
- Vaporization
- Thermal decomposition (ablation)

>100
- Carbonization
- Formation and breaking of vacuols

>300

- Thermoablation of the tissue

Tab. 1.2 : Thermal effects of laser radiation

However, it was observed that not only the temperature achieved but also the temporal duration
of this temperature plays a significant role for the induction of irreversible damage. It is
illustrated in fig. 1.9 how the critical temperature and the corresponding temporal duration
relate to each other if irreversible damage is meant to occur. The curve is derived from several
empirical observation.

Fig. 1.9: Critical temperature for the occurrence of cell necrosis
17
Areas in which the temperature reaches values higher than 60°C are coagulated and irradiated
tissue cells become necrotic. Areas with maximum temperatures less than 60°C are treated
hyperthermically only, and the probability of cells staying alive depends on the duration and
temporal evolution of the temperature obtained. For a quantitative approximation of both the
remaining active molecules and cells at certain temperature level, Arrhenius’ equation is very
useful

O ÷ ÷
A
÷ ÷ =
í
' )
) ' (
exp(
) (
ln
0
0
dt
t RT
E
A
C
t C
t
1.20

where C
0
is the initial concentration of molecules or cells, C(t) is the concentration at a time t,
A is Arrhenius’constant, R is the universal gas constant, and E and are specific tissue
properties.
The local degree of tissue damage is determined by the damage integral given by (1.20) . The
damage degree is defined as the fraction of deactivated molecules or cells given by

) exp( 1
) (
) (
0
0
O ÷ ÷ =
÷
=
C
t C C
t D
d
1.21

Thus, by inserting an appropriate value of the tissue constant , we are able to calculate the
probable damage degree D
d
(t) as a function of time (t). Unfortunately, experimental data for
the two parameters A and E are very difficult to be obtained due to the inhomogeneity of
most tissues and the uncertainty in measuring the surviving fraction. However in literature
[9-13] some values of various tissues are given as the result of different studies.
Laser radiation acts thermally if power densities more than 10 W/cm
2
are applied from either
CW radiation or pulse durations exceeding approximately 1 μs. Typical lasers for coagulation
are Nd:YAG lasers, Dye lasers and diode lasers. CO
2
lasers are very suitable for vaporization
and the precise thermal cutting of tissues. Carbonization and melting can occur with almost any
type of laser if sufficient power densities and exposure durations are provided.
Frequently, not only one but several thermal effects are induced in biological tissues,
depending on the laser parameters. These effects might even range from carbonization at the
tissue surface to hyperthermia a few millimeters inside the tissue.
18
Since the critical temperature for cell necrosis is determined by the exposure time, no well-
defined temperature can be declared which distinguishes reversible from irreversible effects.
Thus, exposure energy, exposure volume, and exposure duration all together determine the
degree and extent of tissue damages. Therefore, careful evaluation of the required laser
parameters is essential. The concomitance of several thermal processes is illustrated in
fig. 1.20. The location and spatial extent of each thermal effect depend on the locally achieved
temperature during and after laser exposure.


Fig. 1.20: Location of thermal effects inside biological tissue










19
1.6.2 Photochemical Interaction

Photochemical interactions are based on empirical observations stating that light can induce
chemical effects and reactions within macromolecules or tissues. These interactions take place
at very low power densities (typically 1 W/cm
2
) and long exposure times ranging from pulses
of seconds to continuous waves. Careful selection of laser parameters yields a radiation
distribution inside the tissue that is determined by scattering. In most cases, wavelengths in the
visible range (e.g. Rhodamine dye lasers at 630 nm) are used because of their efficiency and
their high optical penetration depths. The latter are of importance if deeper tissue structures are
to be reached.
In the field of medical laser physics, photochemical interaction mechanisms play a significant
role during photodynamic therapy (PDT). Frequently, biostimulation is also attributed to
photochemical interactions, although this is not scientifically ascertained.

- Photodynamic therapy (PDT)

During PDT, spectrally adaptaded chromophores are injected into the body. Monochromatic
irradiation may then trigger selective photochemical reactions, resulting in certain biological
transformations. A chromophore compound which is capable of causing light-induced
reactions in other non-absorbing molecules is called a photosensitizer. After resonant excitation
by laser irradiation, the photosensitizer performs several simultaneous or sequential decays
which result in intramolecular transfer reactions. At the end of these diverse reaction channels,
highly cytotoxic reactands are released causing an irreversible oxidation of essential cell
structures.

- Biostimulation

Biostimulation is believed to occur at very low irradiances and to belong to the group of
photochemical interactions. Unfortunately, the term biostimulation has not been scientifically
very well defined, so far. The potential effects of extremely low laser powers (1-5mW) on
biological tissues have been a subject of controversy, since they were first claimed by the
Hungarian surgeon Mester at the end of the 1960s. Wound healing and anti-infiammatory
properties by red or near infrared light sources such as helium-neon lasers or diode lasers were
20
reported. Typical energy fluences lie in the range 1-10 J/cm
2
.
In several cases, researchers have noticed improvements in the patients. But in a few studies
only, results could be verified by independent research groups. Moreover, contradictory results
were obtained in many experiments. However, only very few patients were treated, and no
clinical protocols were established. Furthermore, the success is rather doubtful, since ln many
of these diseases 50 % of the patients are spontaneously cured even without any treatment.
According to Wilder-smith (1988), the distinction from an ordinary “placebo” effect is thus
rather difficult to be verified.
According to Karu (1987), local wound healing effects with helium-neon or diode lasers may
be explained by the action of low-intensity light on o proliferation. In the area of such injuries,
conditions are usually create preventing proliferation such as low oxygen concentration or pH.
The exposure to red or near infrared light might thus serve as a stimulus to increase cell
proliferation. When irradiating fresh wounds the effect of biostimulation is found to be
minimal or even nonexistent. This observation is probably due to the fact that cell proliferation
is very active in fresh wounds, and regeneration is not significantly altered by laser irradiation.
However, biostimulation is a research field with a lot of speculation involved. Usually, the
controversy stems from our inability to specify the photochemical channels of potential
reactions. Detailed investigations in this area and reproducible experimental results are badly
needed.


1.6.3 Photoablation

Photoablation is a laser-tissue interaction where the biological tissue is decomposed when
exposed to high intense laser irradiation. With this effect a direct breaking of molecular bonds
by high-energy (UV) photons is obtained. Typical threshold values of this type of interaction
are 10
7
-10
8
W/cm
2
at laser pulse durations in the nanosecond range. The ablation depth (which
defines the depth of tissue removal per pulse) is determined by the pulse energy up to a certain
saturation limit. The geometry of the ablation pattern itself is defined by the spatial parameters
of the laser beam. The main advantages of this ablation technique lie in the precision of the
etching process, its excellent predictability, and the lack of thermal damage to adjacent tissue.
Today, in medical applications, photoablation is one of the most successful techniques for
refractive corneal surgery, in which the refractive power of the cornea is altered in myopia,
21
hyperopia or astigmatism.
In order to obtain a physical explanation of the photoablation process, let us assume that two
atoms A and B are bound by a common electron. The character of the bonds within an organic
material is primarily covalent. The corresponding energy level diagram of the ground and
several excited states is shown in fig. 1.21

Fig. 1.21: Energy level diagram for photoablation

Due to the macromolecular structure, each electronic level is split into further vibrational states.
Absorption of a photon may promote the two atoms to an excited state (AB)*. Usually, the
excitation is achieved by satisfying the Franck-Condon principle. It states that the radial
distance of the two nuclei of the atoms A and B will not be affected during the process of
excitation due to the small electron mass. Thus, transitions characterized by a vertical line in
the energy level diagram are favored. The probability of such transition is increased if the
maxima of the corresponding Schrödinger function of the initial and final states coincide at the
same radial distance.
If a UV photon is absorbed, the energy gain is usually high enough to access an electronic state
which exceeds the bond energy. In this case, the two atoms A and B may dissociate at the very
next vibration. Thus, photoablation can be summarised as a two-step process :

- excitation: AB + h  (AB)*
- dissociation: (AB)*  A + B + E
k


22
The dissociation energies of some typical chemical bonds, the wavelengths and the
corresponding photon energies of selected laser systems are given in the following tables:













Tab. 1.3 : Dissociation energies of some Tab. 1.4: Laser systems used in
chemical bound medical applications

When comparing both tables, we find that only photons from UV lasers, typically excimer
lasers, provide an energy sufficient for dissociating such bonds. Therefore, the interactions
mechanism of photoablation is limited to the application of UV light.
However, it was observed that a certain threshold intensity must be applied to achieve
photoablatlon. Above this intensity, a well-defined depth is ablated ,depending on the
absorption coefficient and the incident intensity. If the incident Intensity is moderate and such
that the ablation depth is smaller than the corresponding optical absorption length, subsequent
pulses will enter partially irradiated tissues as well as unexposed tissues. Therefore, only the
first few pulses are unique. After these a linear relation between the number of applied pulses
and the total etch depth is obtained. In practice, the etch depths are averaged over several
pulses and noted as ablation depth per pulse. This value is reproducible within an uncertainty
of about 10% for most materials which is an excellent value when taking all the
inhomogeneities of tissues into account.



Type of bond

Dissociation energy (eV)
C=O 7.1
C=C 6.4
O H 4.8
N H 4.1
C O 3.6
C C 3.6
S H 3.5
C N 3.0
C S 2.7
Laser type Wavelenght
(nm)
Photon energy
(eV)
ArF 193 6.4
KrF 248 5.0
XeCl 308 4.0
XeF 351 3.5
Argon ion 514 2.4
Dye - Rhodamine 595 2.2
He–Ne 633 2.0
Diode 800 1.6
Nd:YAG 1064 1.2
Er:YAG 2940 0.4
CO
2
10600 0.1
23
1.6.4 Plasma-induced Ablation

When obtaining power densities exceeding 10
11
W/cm
2
in solids and fluids a phenomenon
called optical breakdown occurs. As result of a optical breakdown a plasma-induced ablation
phenomenon can occur. Sometimes, plasma-induced ablation is also referred to as plasma-
mediated ablation [14-16]. Plasma-induced ablation permit an ablation of the tissue by ionizing
plasma formation. By means of plasma-induced ablation, very clean and well-defined removal
of tissue without evidence of thermal or mechanical damage can be achieved.
The most important parameter of plasma-induced ablation is the local electric field strength E
which determines when optical breakdown is achieved. If E exceeds a certain threshold value,
for istance, if the applied electric field forces the ionization of molecules and atoms,
breakdown occurs. The electric field strength itself is related to the local power density I by
the basic electrodynamic equation


2
0 0 0
2
1
cE I c = 1.22

where
0
c is the dielectric constant, and c is the speed of light. For picosecond pulses, typical
threshold intensities of optical breakdown are 10
11
W/cm
2
. Hence the corresponding electric
field amounts to approximately 10
7
V/cm. This value is comparable to the average atomic or
intramolecular Coulomb electric fields, thus providing the necessary condition for plasma
ionization. Within a few hundred of picoseconds, a very large free electron density with typical
value of 10
18
/cm
3
is created in the focal volume of the laser beam. In general, plasma
generation due to an intense electric field is called dielectric breakdown. The term optical
breakdown especially emphasises that UV, visible and IR light are strongly absorbed by the
plasma. The physical principles of breakdown have been investigated in several theoretical
studies [17-22].
The initiation of plasma generation can be two-fold [23]. It was observed that either
Q-swithced pulses in the picosecond or femtosecond range can induce a localized microplasma.
In Q-swithced pulses, the initial process for the generation of free electrons is supposed to be
thermoionic emission. In mode locked pulses, multi-photon ionization may occur due to the
high electric field induced by the intense laser pulse.
In general, the term multi-photon ionization denotes processes in which coherent absorption of
24
several photons provides the energy needed for ionization. Due to the requirement of coherence,
multi-photon ionization is achievable only during high peak intensities as in picosecond or
femtosecond laser pulses. Plasma energies and plasma temperatures are usually higher in
Q-swltched laser pulses because of the associated increase in threshold energy of plasma
formation. Thus, optical breakdown achieved with nanosecond pulses is often accompanied by
nonionizing side effects.
In either case, however, a few “lucky'' electrons initiate an avalanche effect, leading the
accumulation of free electrons and ions. A free electron absorbs a photon and accelerates. The
accelerated electron collides with another atom and ionizes it resulting in two free electrons.
Each of them will have less individual kinetic energy than the initial electron. Again, these free
electrons may absorb incoming photons, accelerate, strike other atoms, release two more
electrons, and so forth. The basic process of photon absorption and electron acceleration taking
place in the presence of an atom is called inverse Bremsstrahlung. It is based on a free-free
absorption, for istance, a transition where a free electron is present in the initial and final states.
In order to fulfill the conservation laws of energy and momentum, this event must necessarily
take place in the electric field of an ion A
+
or a neutral atom. The process is schematically
represented as :

h + e + A
+
 e + A
+
+ E
k

In particular, the important feature of optical breakdown is that it renders possible an energy
deposition not only in pigmented tissue but also in nominally weakly absorbing media. This is
due to the increased absorption coefficient of the induced plasma. Thereby, the field of
potential medical laser applications is considerably widened. Especially in ophthalmology,
transparent tissues - like cornea and lens - become potential targets of laser surgery. It is also
worthwile noting that during the process there is no restriction of the photon energy involved.
The electron may increase its kinetic energy by means of absorbing any amount of energy, thus
leading to a very short risetime of the free electron density of the order of picoseconds. In order
to achieve optical breakdown the irradiance must be intense enough to cause rapid ionization
so that losses do not quench the electron avalanche. According to Smith and Haught (1966),
inelastic collisions and diffusion of free electrons from the focal volume are the main loss
mechanisms during avalanche ionization.

25
1.6.5 Photodisruption

Photodisruption effect consist on the fragmentation and cutting of tissues by mechanical forces .
In general, the physical effects associated with optical breakdown are plasma formation and
shock wave generation. If breakdown occurs inside soft tissues or fluid, cavitation and jet
formation may additionally take place.
Cavitation is an effect that occurs when focusing the laser beam not on the surface of a tissue
but into the tissue. By means of the high plasma temperature, the focal volume is vaporized
and a bubble, typically made of water vapor and carbon oxides, is formed. As a result of the
outer static pressure, the boubble implodes determining a permanent tissue damage.
Jet formation occur when cavitation bubble collapse in the vicinity of a solid boundary. In this
case, a high speed liquid jet directed toward the wall is produced. if the bubble is in direct
contact with the solid boundary during its collapse, the jet can cause high-impact pressure
against the wall resulting in a severe damage and erosion of tissue. Thus, consequent to jet
formation phenomenon, bubbles attached to solids have the largest damage potential.
When discussing plasma-induced ablation, any secondary effects of the plasma have been
neglected. At higher pulse energies, and thus higher plasma energy, shock waves and other
mechanical side effects become more significant and might even determine the globe effect
upon the tissue. Primarily, this is due to the fact that mechanical effects scale linearly with the
absorbed energy. Then, because of the mechanical impact, the term photodisruption is more
appropriate.
During photodisruption, the tissue is split by mechanical forces. Whereas plasma-induced
ablation is spatially confined to the breakdown region, shock wave and cavitation effects
propagate into adjacent tissues, thus limiting the localizability of the interaction zone. For
pulse durations in the nanosecond range, the spatial extent of the mechanical effects is already
of the order of millimeters even at the very threshold of breakdown. Actually, purely plasma-
induced ablation is not observed for nanosecond pulses, because the threshold energy density
of optical breakdown is higher compared to picosecond pulses, and the pressure gradient scales
with plasma energy.
Hence, for nanosecond pulses, optical breakdown is always associated with shock wave
formation even at the very threshold. Since adjacent tissues can be damaged by disruptive
forces, the presence of these effects is often an undesired but associated symptom. In contrast,
picosecond or femtosecond pulses permit the generation of high peak intensities with
26
considerably lower pulse energies. With these extremely short pulse durations, optical
breakdown may still be achieved with significantly reducing plasma energy and, thus,
disruptive effects. Moreover, spatial confinement and predictability of the laser-tissue
interaction is strongly enhanced. Since both interaction mechanisms, plasma-induced ablation
as well as photodisruption, rely on plasma generation, it is not always easy to distinguish
between these two processes. Actually, in the 1970s and 1980s, all tissue effects evoked by
ultrashort laser pulses were attributed to photodisruption. It was only because of recent
research that a differentiation between ablations solely due to ionization and ablations owing to
mechanical forces seems justified.
In general, photodisruption may be regarded as a multi-cause mechanical effect starting with
optical breakdown. In fig. 1.22, a schematic sequence of these processes is illustrated
indicating their relations to each other. Moreover, the distinction between photodisruption and
plasma-induced ablation is emphasized


Fig. 1.22: Scheme of the physical process associated with optical breakdown.Percentagesgiven
are rough estimates of the approximate energy transferred to each effect


The four effects, plasma formation, shock wave generation, cavitation, and jet formation , all
take place at a different time scale. Plasma formation begins during the laser pulse and last for
a few nanoseconds afterwards. This basically is the time needed by the free electrons to diffuse
into the surrounding medium. Shock wave generation is associated with the expansion of the
plasma and, thus, already starts during plasma formation. However, the shock wave then
27
propagates into adjacent tissue and leaves the focal volume. Approximately 30-50 ns later, it
has slowed down to on ordinary acoustic wave.
Cavitation, finally, is a macroscopic effect starting rougly 50-150 ns after the laser pulse. The
time delay is caused by the material during the process of vaporization. Usually, the cavitation
bubble performs several oscillations of expansion and collapses within a period of a few
hundred microseconds. Since the pressure inside the bubble again increases during collapse,
each rebound of the cavitation bubble is accompanied by another shock wave. Furthermore,
every collapse can induce a jet formation if the bubble is generated in the vicinity of a solid
boundary.
Photodisruption has become a well-established tool of minimally invalid surgery (MIS), since
it was introduced by Krasnov (1973) and then further investigated by Aron-Rosa et al (1980)
and Fankhauser et al (1981) [24-26]. Two of the most important applications of
photodisruptive Interaction are posterior capsulotomy of the lens, frequently being necessary
after cataract surgery, and laser-induced lithotrispy of urinary calculi.




1.7 Laser Applications in the medical field

Since their first appearance, lasers have found numerous applications in the field of medicine.
The first medical application of lasers dates back to 1962, when a ruby laser was used for
reattachment of the retina in the eye. It was the invention of the CO
2
laser in 1967 with its high
absorption by water that opened up the vast field of surgical applications.
After 1974, the availability of optic fibers permitted the combination of laser technology with
endoscopy. In table 1.4 the most used laser systems in medical applications are listed.
The high absorption rate in water, hemoglobin, and melanine guarantees efficient clinical
applicatios with optimal medical results.
Nowadays, the development of new technologies and the increasing understanding of the
utilization of that technology by the medical community have brought a continuous increase of
laser use in a wide range of medical applications. Below, some of the principal applications of
laser in modern medicine are described .
28
- Laser in Dermatology

In dermatology, thermal effects of laser radiation are commonly used, especially coagulation
and vaporization [27, 28]. Since the optical parameter of skin, i.e. absorption and scattering, are
stronghly wavelength-dependent, various kinds of tissue reaction can be induced by different
laser systems. In clinical practice, five types of lasers are currently being used: argon ion lasers,
dye lasers, CO
2
lasers, Erbium lasers and Nd:YAG lasers.
Radiation from argon ion laser is stronghly absorbed by hemoglobin and melanin. This laser is
thus predestined for superficial treatments of highly vascularized skin. One of the most
frequent applications of this laser is upon patients with plane angioma pathologies
The first laser to produce selective light induced injury was the pulsed dye laser. This device is
used to treat a variety of blood vessel abnormality and, in particular, in the last years is
prefered to the argon laser for treatment on port wine stains (plane angiomas). The pulsed dye
laser approved by FDA in 1988 has revolutionized the treatment of this disorder and has them
been used to treat other disorders of the blood vessel including broken blood vessels on the
face, spider veins on the legs, different types of scars and even warts [29-33]. Pigmented laser
disorders such as aging spots, café-au-lait spots, brown birthmarks and tattoos are treated by
pulsed laser technology [34, 35]. Laser has recently been developed for hair removal by
selectively damaging the hair bulb. The most recent development that generated tremendous
interests are the use of pulsed CO
2
and Erbium lasers for skin resurfacing. By using short
pulses of light that are absorbed by the upper layers of skin, these lasers resurface the skin
making it much smoother and restoring it to a more youthful appearance This treatment is
effective in rejuvenating skin of patients with wrinkles or other sun-induced damage and also
for smoothing the textural change of acne scarring. Compared to chemical peels, laser
resurfacing is more precise. The results are more impressive and side effects are no more
frequent. These dramatic developments have reshaped the ability to treat skin disorders.
Radiation from Nd:YAG lasers is significantly less scattered and absorbed in skin than
radiation from the other lasers. The optical penetration depth of Nd:YAG lasers radiation is
thus much larger. Hence, major indications for this laser in dermatology are given by deeply
located hemangiomas or semimalignant skin tumor.



29
- Laser in Dentistry

Altough dentistry was the second medical discipline (after ophthalmology) where lasers were
applied, it basically remains a field of research. Especially in caries therapy (the most frequent
dental surgery) conventional mechanical drills are still superior compared to most types of
lasers, particularly CW or long-pulse lasers [36, 37]. Only laser systems capable of providing
ultrashort pulses might be an alternative to mechanical drills. However, many clinical studies
and exstensive engineering effort still remain to be done in order to achieve satisfactory results.
Recently, studies have established the disturbing link between periodontal disease, heart
disease and strokes. Those patients with infected gums and gum diseases suffer twice the heart
attack rate and almost triple the stroke rate of those who do suffer from gum infection. The last
one is crucial when arterial cardiovascular damage is present in a patient. Pre-surgical laser
protocol can help eliminate bacteremia by efficiently addressing buildup of bacterial colonies
that cause and sustain periodontal disease in a patient ready to receive open-heart surgery.


- Laser in Otolaryngology

Otolaryngology was one of the first areas in which lasers have been applied. Particularly, the
CO
2
laser has found use in the treatment of airway in which precision, homeostasis and lack of
swelling are necessary. This allows removal of diseased tissue without compromising the
patient’s breathing. Carbon dioxide lasers have been used to remove both benign and
malignant conditions of the larynx (voice box), pharynx (throat) and oral cavity. Similarly,
ND:YAG and KTP lasers can be transmitted through a fiber optic cable to remove growths in
the tracheobronchial tree as well as esophagus. These two wavelengths have been applied to
surgery within the nasal passages and sinuses. The CO
2
and Argon laser have been used to
treat otologic disorders such as otosclerosis. The precision afforded allows delicate removal of
bone tissue and replacement with artificial prothesis to improve hearing. Currently
otolaryngologists are at the forefront of using various imaging modalities to view laser tissue
interaction deep within the head and neck to afford greater therapy precision in the care of
patients.


30
- Laser in Urology

Various types of lasers are utilized to successfully treat a number of diseases on the urinary and
genital systems [38]. The more common diseases include: urinary stones, urothelial tumors,
obstructions due to narrowing of ureteropelvic junction, ureteral structure, benign enlargement
of the prostate urethral structures, benign condylomas and superficial cancer of the penis.
Majority of such treatments require general anesthesia but can be performed as an outpatient
basis with little or no blood loss and excellent results.



- Laser in Podiatry

Many conditions treated by laser include: fungal nails, ingrown nails, deformed nails,
matrixectomies, granulomas, porokeratosis, hemangiomas, pigmented villondular, synovitis,
neurofibromatosis, periungual fibromas, remodeling hypertrophic scars/keloids, adjunct ulcer
therapy (like debridement), plantar warts, skin fissures, tumors, various types of cysts
including foreign body cysts, incision and excision of neuromas and release of sublesional
hematomas.



- Laser in Ophthalmology

In ophthalmology, various types of lasers are being applied today for either diagnostic or
therapeutic purpose [39, 40]. However, lasers have three main applications in ophthalmology:
1) They are used to treat glaucoma, a group of diseases caused by increased pressure in the eye.
2) Refractive surgeons use excimer lasers to reshape the cornea and help the eye focus properly.
Laser vision correction to treat near-sightedness, farsightedness and astigmatism
3) Retina specialists use lasers (argon or ruby lasers) to treat macular degeneration.
The procedure combines a light activated drug with a low-power red light to destroy
abnormal blood vessels.

31
- Lasers in Neuro-surgery

Microsurgical lasers are used for precision cutting, for making incisions into the brain and
spinal cord producing discrete results for the relief of pain. The CO
2
, KTP and ND:YAG are
used frequently in the vaporization and coagulation of tumors, particularly, if located
intracranially or deep within the brain. The use of laser has been particularly beneficial for
very narrow approaches that are required to access tumors in these locations. Fiber-optically
delivered lasers are being more frequently used for the vaporization and shrinkage of spinal
discs to treat patients with herniated discs within the upper, mid and lowerback through
minimally invasive surgical approach using the endoscopic guidance.























32
Chapter 2

INFRARED IMAGING: Principles and applications

2.1 Introduction

Among the different non-destructive techniques that are in use nowadays, Infrared
Thermography (IT) stands as an attractive tool for non-contact inspections [41, 42].
It is one among the numerous techniques used to “see the unseen''. As the name implies, it uses
the distribution (suffix –graphy) of surface temperatures (prefix thermo-) to assess the structure
or behavior of what is under the surface.
Infrared Thermography is a contactless imaging technique with distinct advantages.
It is based on the recording of the natural thermal radiation emission of a body by employing a
infrared camera. The result is the temperature distribution of the body’s surface under
investigation. By analysis and images elaboration it is possible to achieve information about
the condition of surface and subsurface structure of the sample such as:

- Homogeneity degree
- Individuation and characterization of anomalies
- Monitoring of the sample under external heating

In general infrared technique can be deployed in two different experimental approches
depending on the specific application: passive and active method.
Passive thermography is based on the detection of the infrared images under natural conditions
without an external stimulation. In active thermography a sequence of infrared images is
detected during or after application of an external thermal stimulation.
Distinctions between the passive and active approaches are not as sharp as it seems. For
example, although maintenance investigation of buildings is generally considered as a passive
thermography application, it can be performed by the active approach as well. As it is known,
water in a wall activates undesired chemical reactions in building and thus a degradation
process. In the case of monumental buildings, due to the extremely high cost of restoration, it is
first necessary to assess the moisture problem. This can be done by heating the wall uniformly
33
from one side while observing the isotherm pattern on the other side. The thermal map
recorded depends on the water content since the heat capacity is higher for water than for dry
material. lt is interesting to note that moisture evaluation of buildings and roofs in particular
has been reported by passive methods as well. In this case, thermographic inspection is
performed at night or early in the morning and detection is based on the fact that moist areas
retained day sunshine heat better than did sound dry areas because of the high thermal capacity
of water. ln that case, too, the passive label is debatable since solar heating is a form of external
thermal stimulation.
What is generally recognized is that if one does not need thermal stimulation, the procedure is
called passive and active otherwise.
ln this chapter both approaches are described, focusing the attention on different active
procedures. The most important features of infrared detectors are discussed and, finally, the
chapter is completed with an overview on the new perspectives of infrared imaging
applications in medical science.



















34
2.2 Passive thermography

The first law of thermodynamics concerns with the principle of energy conservation and states
that an important quantity of heat is released by any process consuming energy because of the
law of entropy. Temperature is thus an essential parameter to measure in order to assess proper
operation. In passive thermography, the sample is observed under natural thermal condition
while no external thermal perturbation is applied. In this procedure, abnormal temperature
profiles indicate a potential problem, and a key term is the temperature shift with respect to a
reference, often referred to as the delta-T ( T) value or the hot spot. A T value of a few
degrees (> 3 °C) is generally suspicious, while higher values indicate strong evidence of
abnormal behavior. Applications such as control of industrial processes [43-45], building
inspection [46-49], military surveys, welding, fire detection (for example in the forest), and
some medical investigation make use of this mode of operation, where surface temperature
distribution, as it is, contains relevant information concerning possible disorders, the presence
or absence of anomalies, and so on.

Fig. 2.1: Passive thermography on a sclerodermic patient: in the image
hot spots associated to some morphea plaques are well visible

Generally, passive thermography is rather qualitative since the goal is simply to individualize
anomalies. However, if thermal modeling is available, some investigations provide quantitative
measurements, so that measured surface temperature (isotherm) can be related to specific
behaviors or subsurface flaws. For instance, such dedicated modeling helps us to understand
needle heating during high-speed sewing in the automobile industry. This makes it possible to
optimise sewing operations through needle redesign and needle cooling, with significant
economic and quality benefits, due to the millions of products sewed daily (seat cushions,
airbags, etc…) [50].
35
2.3 Active thermography : Testing procedures

ln active thermography it is necessary to transfer some energy to the specimen inspected in
order to obtain significant temperature differences emphasizing the presence of subsurface
anomalies.
Defect detection principle, in active thermography, is based on the followed idea:

If a thermal front, travelling inside the sample, meet a dishomogeneous area, its part is
reflected behind and modifies the temperature of the surface over the defect

So, from the acquisition of a sequence of superficial images, during and after a thermal
stimulation, it is possible, by images elaboration, to obtain information about structural
subsurface homogeneity, defect individuation and quantification purposes (inverse problem).





Fig. 2.2: Active thermography applied on a wood specimen: a) visible image of the specimen, b) contrast thermal
image obtained after a procedure of pulsed thermography. The thermal contrast image emphasizes the
presence of air holes to different depth inside the wood


Hence, contrary to the passive approach, the active approach requires an external appropriate
heat source to stimulate the materials for inspection.
This is a crucial aspect since lack of success in active termography applications is often related
to problems with the thermal stimulatlon device, such as the lack of homogeneity of the
heating source. ln pulsed thermography, when step heating is used to reveal subsurface defects,
it is necessary to generate a thermal transient inside the part. Such a thermal transient can be
ACTIVE
THERMOGRAPHY
a b
36
generated as either a static or a mobile configuration. Such a thermal stimulation can be either
hot (hotter than the sample) or cold (cooler than the sample) since a thermal front propagates in
the same way in both situations. The cold approach has, under certain circumstances, several
advantages over the warm approach: for example, when the part to be inspected is already hot,
like the external part of a curing oven in the case of graphite-epoxy panels. In these cases, it is
uneconomical adding extra heat to the part, while cooling is more economical. Another benefit
is that a cool thermal stimulating device will not generate any thermal noise since it is at a
reduced temperature with respect to the component to be inspected. Such a cool thermal
perturbation device can be deployed using :

- Cool water jets
- Cool air/gas jets
- Snow or ice (with contact)
- Cool water bag (with contact)

More traditionally, a hot thermal perturbation is used, which can be:

- High-powered cinematographic lamps
- Sequence of quartz infrared line lamps
- High-powered photographic flashes
- laser beam (Scanned or single beam)
- Heat gun, hot water jets, hot air jets, hot water bag

The main desirable characteristics of the thermal perturbation (either cool or warm) is
repeatability, uniformity, timeliness, and duration.
Repeatability is needed in order to compare among results obtained on many identical parts. It
is worthwhile to notice that lamp aging can reduce power deposit significatively: up to 40% of
light intensity can be lost due to the deposition of evaporated tungsten from the filament on the
inner lamp glass envelope.
Uniformity is necessary when inspection is performed on wide surfaces. Dishomogeneous
thermal stimulation will produce spurious hot (cold) spots, which may be interpreted as
subsurface flaws (especially in the case of the pulsed scheme, described below). This
uniformity characteristic is hard to achieve. However, specific deconvolution algorithms,
37
specific calibration and contrast computations can allow for stimulation correction after its
application.
For the pulsed scheme, timeliness and duration are other important issues for thermal
stimulation. In this case, in fact, for a better agreement with theoretical models and to enable
data inversion, a square thermal pulse is needed. In order to optimize this parameter a
mechanical shutter may be used in front of the thermal perturbation device for the start-stop of
the pulse. Nowadays, several active thermographic techniques exist differing from each other
mainly in the way data is acquired and/or processed as reviewed and discussed briefly below.
In this description more emphasis is given to the presentation of the pulsed termography
method because part of the experimetal procedure used in this thesis work (described in the
next chapters) make use of this active technique.



2.3.1 Pulsed Thermography


Pulsed thermography (PT) is one of the most common thermal stimulation methods used in
infrared imaging [51-62]. One reason for this is how fast is the inspection, during which a short
thermal stimulation pulse lasting from a few milliseconds for high-conductivity material (such
as metal) to a few seconds for low-conductivity samples (such as plastics, wood, graphite-
epoxy laminates) is used. Basically, PT consists of a fast heating of the sample under
investigation and then recording the temperature decay curve. Qualitatively, the phenomenon
can be described in the following. The temperature of the material changes rapidly after the
initial thermal pulse because the thermal front propagates by diffusion under the surface and
also because of radiation and convection losses. The presence of a subsurface defect modifies
the diffusion rate. Thus when observing the surface temperature, a different temperature with
respect to the surrounding sound area appears over a sub-surface defect once the thermal front
has reached it. Consequently, deeper defects will be observed later and with reduced contrast.
In fact, the observation time t is a function (in a first approximation) of the square of the depth
z [63] and the loss of contrast c is inversely proportional to the cube of the depth [64]


o
2
z
t ~

3
1
z
c ~
2.1
38
where is the thermal diffusivity of the material. These relations (2.1) show two limitations of
PT : observable defects will generally be shallow and the contrast weaks.
A well known empirical rule says that: the radius of the smallest detectable defect should be at
least one to two times larger than its depth under the surface [65].
In general, there are two different operation way in PT. The sample is heated from one side
while thermal data is collected either from the same side ( i.e. reflection mode) or from the
opposite side (i.e. transmission mode) (Fig. 2.3).



Fig. 2.3: Pulsed Thermography set-up: reflection and trasmission mode

Reflection is used when inspecting defects closer to the heated surface, whilst transmission is
preferred for detecting defects closer to the non-heated surface (i.e. deeper defects). In general,
reflection is most used because resolution is higher than trasmission mode and it is easier to be
employed since both sides of the sample do not need to be available. Although deeper defects
can be detected in transmission, depth information is loss since thermal waves will travel the
same distance whether their strength is reduced by the presence of a defect or not. Hence, depth
quantification is not possible in transmission.
Defective zones will appear at higher or lower temperature with respect to nondefective zones
on the surface, depending on the thermal properties of both the material and the defect.
It is now interesting to make the link with thermal wave theory. Mathematically, a pulse can
be decomposed into a multitude of individual sinusoidal components. In that respect, pulse
heating of a sample corresponds to the simultaneous launching into the sample of thermal
waves of various amplitudes and frequencies, in a transient mode. Utilization of a thermal
pulse for stimulation of the sample is thus practical since thermal waves of many different
frequencies are tested simultaneously (the flat frequency spectrum of a Dirac pulse).
39
The detection phenomenon in PT is illustrated in Fig. 2.4


Fig. 2.4: Temperature evolution curve after absorption of a rectangular pulse: 1) sample made
of homogeneous material; 2) sample containing a subsurface flaw

the surface temperature evolution following the application of the initial heat pulse is plotted
for a thin homogeneous plate (curve 1) and for the same plate when a delamination (air layer)
is present under the surface (curve 2). In the logarithmic scale we notice that the central portion
of the temperature decay follows a slope of -1/2. In fact, for a semi-infinite medium, after
absorption of a Dirac pulse, this temperature decay conforms to [66]

t e
Q
T
f
t
= A
2.2


where T is the temperature increase of the surface, Q the quantity of energy absorbed, and
e
f
=(k C)
1/2
is the thermal effusiviy of the material, with k being the thermal conductivity, the
mass density, C the specific heat, and t the time, respectively.
In PT the short heating employed prevents damage to the component. Heating is generally
limited to a few degrees above the initial sample temperature. Obviously, the stimulation
source must be nondestructive and thus it must not damage the inspected surface either
chemically or physically: this will limit its strength. This can be an issue when high-intensity
pulses are employed, especially when high-intensity heating is directed on delicate surfaces
(e.g. frescoes, paintings of old Masters).
40
It is worthwhile to notice that a cold front (heat removal) will propagate in the same way as a
hot front (heat deposit) inside the material. ln some inspection situations, this cold approach is
cheaper: for instance, in case of the inspection of a component already at a high temperature
(with respect to ambient temperature) due to the manufactured process such as for the cuing of
bonded parts. Moreover, it may be attractive to preheat (if necessary) the surface with a low
rate of energy deposition (e.g. by means of electric heating wires). This approach avoids
relying on high-intensity local heat sources, which are more costly, susceptible to rapid aging,
and potentially dangerous for the surface because of the high thermal stress they induce. Cold
thermal sources are also advantageous since they do not generate thermal reflective noise,
which can perturb the measurement.
As for the detection depth this is limited since termography is a border technique. Anyway the
anomalies (such as cracks, etc ) usually start close to the surface.
In the next section more information about a complete thermogram sequence are given. The
study of the temporal temperature profile is the core in the data-elaboration in the pulsed
thermography method. This is, in fact, the basis for defect detection in active thermography as
described next.



2.3.1.1 The complete thermogram sequence


In pulsed termography each infrared images sequence recorded, before and after external
thermal stimulation, can be analized pixel to pixel. The temporal evolution associated to every
pixel follows a rate that can be separated in different step.
In general, the complete thermal temporal sequence for a single pixel is composed of 5
distinctive elements depicted in fig. 2.5. At time t
0
, which defines the time immediately before
heat reaches the sample’s surface, a cold image (1) is captured. The cold image can be used to
eliminate spurious reflections due to emissivity variations and to reduce fixed pattern artifacts.
This is attractive for thermal data visualization and quantification although it is less useful
when working with phase delay images. During (and shortly after) the application of a heat
pulse, the acquired thermograms are temperature saturated (2), i.e. the reading is out of the
calibration scale and no accurate measure can be computed. The actual number of saturated
thermograms depends on the sampling frequency and on the thermal properties of the material
41
under inspection: low conductivity materials saturates longer than high conductivity materials,
thus more thermograms, saturated or not, will be recorded using high sampling rates. Saturated
thermograms give no valuable information and therefore can be safely discarded from the
processing stage.











Fig. 2.5 : Complete thermogram sequence of a single pixel: 1- cold image, 2- saturated thermograms,
3- ERT-Early Recorded Thermogram, 4- thermogram sequence and 5- LRT- Last Recorded Thermogram

The first useful thermogram that comes into sight after saturation is known as the Early
Recorded Thermogram (ERT) (3). Ideally, defects are still not visible on the ERT. Anyhow,
this condition is not always encountered in practice, especially when inspecting shallow defects
on high conductivity materials using low sampling frequencies and/or when strong
non-uniform heating is present. Normally, this situation does not constitute a problem for
defect detection purposes. However, since depth is a function of time (z ~ t
1/2
[42]), special
care must be taken in order to perform quantitative analysis. Starting at the ERT at t
1
, all
subsequent thermograms are of interest for defect inspection and constitute the thermogram
sequence (4). The last acquired image at t
N
(5) corresponds to the Last Recorded Thermogram
(LRT). From this point, temperature variations are considered negligible. Deviation from the
t
1/2
dependency on the useful part of the thermogram provides an indication of the presence of
a defective area (see fig. 2.4). In order to obtain information about the presence of anomalies or
defects an important parameter obtainable from an infrared sequence is the thermal contrast
( T). There are different definitions of thermal contrast in literature, the most used of them are
described in the next section .
TEMPORAL
EVOLUTION
42
2.3.1.2 Definitions of thermal contrast

Several inversion procedures for defect characterization, (i.e. determination of the size, depth
and thermal resistance of a defect, or for the evaluation of surface coatings) make use of
normalized thermal contrast as an input signal. During real experiments, the thermal contrast is
quite corrupted by non-uniform heating, non-uniform emissivity and absorption distributions of
the inspected sample. There are some classical definitions for the thermal contrast which are
acknowledged to be quite effective in reducing the effects of non uniform heating and non
uniform optical properties [67, 68]. Each contrast definition is applied depending on the
situation. In the following these definitions are presented. Then, at the end of the section, a
definition of a newer contrast, the DAC contrast, that is reference-independent and very
effective in reducing non-uniform heating effects, is given [69, 70].


- Contrast for the transmission mode

After the absorption of the heating energy by the front surface of the sample, heat diffusion
within the material yields thermograms (temperature T versus time t curves) whose qualitative
shapes for the rear surface are shown in Fig. 2.6 . T
SA
(t) is the thermogram in a sane area.
while T(t) is the thermogram in a defective area. The temperature in the defective area shows a
slower increase than in the sane area. The difference of the two thermograms leads to the
thermal contrast on the rear surface T
TRANS.


Fig. 2.6 : Temperatre and contrast for the transmission mode

43
The curves in fig. 2.6 correspond to a theoretical case of uniform heating and adiabatic heat
transfer assumption. In real life, heating is non uniform and specimens are subject to heat
losses, which yields thermograms on both sane and defective areas showing distinct maxima.
To overcome the effect of non uniform heating, a classical definition of the contrast [67]
suggests to determine the contrast by first normalizing each thermogram on its peak value, then
to calculate the difference between the two normalized thermograms, which leads to:


MAX SA
SA
MAX
T
T
T
T
T
,
÷ = A
2.3

where Sa = Sane area and MAX = maximum of thermogram. During the processing of the
infrared sequences obtained from the PT experiments, eq. 2.3 is performed for each pixel of
the images. The local normalization aspect of each thermogram used in eq. 2.3 yields an
effective attenuation of the non uniformity of the heating.


- Contrast-for the reflection mode

After the heat excitation, heat diffusion within the material yields thermograms whose
qualitative shapes for the front surface are shown in fig. 2.7. T
SA
(t) is the thermogram in a sane
area, while T(t) is the thermogram in a defective area. We can see that the defective area shows
a slower decrease than the sane area. The difference of the two thermograms leads to the
thermal contrast on the front surface T
REF


Fig. 2.7 : Temperature and contrast for the reflection mode

44
|
|
.
|

\
|
÷ = A
ERT SA
SA
ERT
SF
SHORT
T
T
T
T
t
T
,
1
t
Similarly to the transmission case, there is a definition of a locally normalized contrast. Such a
definition is convenient for the reflection mode when a stabilization temperature is attained (e.g.
high conductivity materials). The curves shown in fig. 2.7 are obtained under the assumption of
uniform heating and negligible heat losses. In real experiments, heating is not uniform which
leads to different asymptotic levels for both defective and sane area thermograms. To reduce
the effect of non uniform heating, each thermogram is normalized on its asymptotic level, and
then the difference between the two normalized thermograms is determined, which leads to
LRT SA
SA
LRT
LONG
T
T
T
T
T
,
÷ = A
2.4

where LONG = normalization at long times, Sa = Sane area and LRT = last recorded
temperature. For poor heat conductor materials, the asymptotic level at the end of the
thermograms as shown in fig. 2.7 does not exist, and eq. (2.4) cannot be used. In such
situations, the asymptotic adiabatic temperature can be evaluated from the thermograms at a
short time (i.e. just after the flash heating). Indeed, the thermograms at those short times follow
the modelling of a semi-infinite sane medium. If we denote as T
ERT
the temperature recorded
immediately after the flash heating at time t
S
of the first frame, for example, we have :

( )
2 / 1
SF ERT
t T T t =
·
2.5

where t
SF
is the Fourier number corresponding to time t
S
(i.e. t
SF
=a
Z
t
S
/e
2
, where a
Z
is the
thermal conductivity in the z direction). The evaluation of the asymptotic level T
0
for the sane
area is performed the same way starting from temperature T
0S
at time t
S
. The locally
normalized contrast for each pixel can therefore be calculated by :

2.6



Where SHORT = normalization at short times, Sa = Sane area and LRT = early recorded
temperature. T
SHORT
is convenient for the reflection mode when the thermal profiles do not
reach stabilization (e.g. low conductivity materials and/or short acquisition times).
45
t e
Q
T
f
t
= A
- Differential Absolute Contrast (DAC)

In the last two sections it has been presented the thermal contrast definitions for the
transmission and the reflection modes (for long and short times), respectively. Although the
impact of non-uniform heating is minimized through the normalization process, the exact
location of a non-defective area is still needed to compute the contrast. It is also possible to use
other thermal contrast definitions such as the differential absolute contrast (DAC) which has
proved to effectively reduce the effect of non-uniform heating and which does not require the
definition of a sane area [69, 70].
The DAC technique is based on the absolute thermal contrast definition:

) ( ) ( ) ( t T t T t T
SA
÷ = A
2.7

and the 1D solution of the Fourier diffusion equation for a Dirac pulse applied on the surface
(z=0) of a semi-infinite body, which may be written as follows [66]

2.8


where T(t) is the temperature evolution, Q(J/m
2
) is the injected energy at the surface, and
e
f
=(k C)
1/2
is the thermal effusivity of the sample. The last one is a thermophysical property
relevant to transient surface heating processes measuring the material ability to exchange heat
with its surroundings.
In the DAC method, instead of looking for a non-defective area, T
SA
is computed locally
assuming that there is no defect visible on the first few images (i.e. it behaves as a sane area).
The first step is to define t' as a given time value between the instant when the pulse is
launched and the precise time when the first defective spot appears on the thermogram
sequence, i.e. when there is enough contrast for the defect to be detected.
At t', there is no indication of the existence of a defective zone yet, therefore the local
temperature for a sane area is exactly the same as for a defective area:

46

'
) ' ( ) ' (
t e
Q
t T t T
f
SA
t
= =

¬

' ) ' ( t t T
e
Q
f
t =
2.9

From the above equation, T
SA
can be computed for every pixel at time t, and substituting it into
the absolute contrast definition, eq. 2.7, lt follows that :
) ' (
'
) ( t T
t
t
t T T
DAC
÷ = A
2.10

Eq. 2.10 needs to be normalized. This can be done in a similar way as described for the
contrast in the reflection mode. The temperature recorded immediately after the flash heating
T
ERT
at time t
S
is given by eq. 2.5. From this definition and eq. 2.10, the normalized
temperature for a defect is given by :

S F
ERT
t
T
t T
T
t T
t
1 ) ( ) (
=
·
2.11

and for a sound area:

S F
S
S F
S
S F S A ERT
S A
S A
t
t
t
t t T
t
t
t T
t
t
t T
t T
t
t
T
t T
t
t
t
1
) ' (
'
) ' (
'
) ' (
'
) (
,
,
= = =
·
2.12

From eq. 2.11 and eq. 2.12, we obtain a normalized definition for the DAC contrast:


SF
S
S ERT
N DAC
t
t
t
t T
t T
T
t
1
) (
) (

÷ = A
2.13

47
The DAC method has proven to be very efficient in reducing artifact from non-uniform heating
and surface geometry [71]. An interactive step is involved in the process of determining t', for
which semi-automated and fully automated algorithms [70] are available. Research is now
moving towards the study of a modified DAC approach. This last technique utilises thermal
quadrupoles and the Laplace transform in order to derive a solution that includes the plate
thickness.





2.3.2 Step Heating (Long Pulse)


In contrast to the thermal stimulation scheme, for which the temperature decay is of interest, in
case of Step Heating the increase in surface temperature is monitored during application of a
stepped heating pulse (the sample is continuously heated, at low power). Variations of surface
temperature with time are related to sample features as in pulsed thermography. This technique
of stepped heating (SH) is sometimes referred to as time-resolved infrared radiometry (TRIR).
Although strictly speaking, techniques using pulsed heating could also be considered time-
resolved. Time-resolved means that the temperature is monitored as it evolves during and after
the heating process. Stepped-heating describes the functional dependence of the heating source.
SH has numerous applications, such as for coating thickness evaluation (including multilayered
coatings, ceramics), integrity of the coating-substrate, bond determination or evaluation of
composite structures and characterization of airframe hidden corrosion, among others.
Experimental setup for SH expedients is similar to PT setup (fig.2.3). When SH was developed,
only infrared scanners were available. At that time, one of the main advantages of SH was the
much faster temporal resolution achievable with respect to full-field imaging at a video rate of
about 25 or 30 images/s. This was because in these cases, the vertical mirror of the scanner was
stopped so that the device operated only in line scan mode. More recently, thanks to the
availability of fast FPA-IR cameras, the technique has been simplified. The start of the heating
pulse is triggered with respect to acquisition of infrared images which are recorded sequentially
for later processing of the entire heating period. Data processing includes evaluation of the
thermal transit time t
T
, that is, the time the normalized temperature of a given point in the field
48
of view differs from the evolution obtained in the case of a thick sound specimen. Transit time
t
T
is related to defect depth. fig. 2.8 shows the effect of the coating thickness on the sample
temperature.


Fig. 2.8: Example of curves temperature versus t
1/2
for a sample with a coating
of different thickness for a step heating pulse of duration 1 s.

Advantages of this approach are that defect depth and thermal properties are accessible from a
single measurement without requiring knowledge about a defect-free region (such as to
compute the contrast in PT). In fact, the early time, before the manifestation of subsurface
defects, allows us to calibrate the image in order to take into account spatial variations of
emissivity and reflectivity.
Although from the experimental point of view SP is different than PT, from a mathematical
point of view, both procedures provide thermal signals that contain exactly the same
information. The choice to rely on PT or SP will thus depend on the applications, and more
specifically, on the time scale of the phenomena of interest: slow phenomena are better
explored with SH and fast phenomena with PT. In general, let be called t
dist
the period of time
required for a given defect to distinguish itself from the background, then the procedure (SH or
PT) that enables to put a maximum energy Q in the specimen during t
dist
will produce the
largest signal (eq. 2.2). It has been suggested [72] that for a given application, a pulse length
slightly smaller than t
dist
can be chosen and the pulse strength can be adjusted so that the
maximum surface temperature reached does not damage the specimen. However, other
practical criteria, such as limitation of power for continuous sources or limitation in the
adjustability of pulse duration for pulsed sources, have to be considered as well.

49
2.3.3 Lock-in Thermography


Lock-in thermography (LT), sometimes called phototherapy radiometry, is based on thermal
waves generated inside a specimen and detected remotely [73-77]. Wave generation, for
example, is performed by periodic deposition of heat on a sample's surface (e.g . through sine-
modulated lamp heating) while the resulting oscillating temperature field, in the stationary
regime, can be recorded remotely through its thermal infrared emission by an IR camera. Lock-
in refers to the necessity to monitor the exact time dependence between the output signal and
the reference input signal (i.e. the modulated heating). This is done with a locking amplifier in
point-by-point laser heating or by computer in full-field employment (lamp) so that both phase
and magnitude images become available.
More specifically, measurement of temperature evolution over the specimen surface permits us
to reconstruct the thermal wave (travelling inside the sample) and to establish values of A
(wave magnitude) and (phase) from four equidistant temperature data points.
Although four points are enough to compute A and , more points allow us to reduce noise
associated with the process. Computer-based commercial systems of LT (such as the lock-in
option of FLIR/CEDIP) works on all available points recorded over one or more modulation
cycles. Such a system thus provides three images: phase , amplitude A, and conventional
thermography image T. The thermographic image is a mapping of the thermal infrared power
emitted, the phase image is related to the propagation time and the modulation image is related
to the thermal diffusivity. Obviously, one of the strong points of LT is the phase image, which
is relatively independent of local optical and infrared surface features and hence it is interesting
for nondestructive-testing purposes. For example, optical features refer to nonuniform heating,
while infrared features may concern with a variable surface emissivity. The depth range of
images is inversely proportional to the modulation frequency of the input signal, so that higher
modulation frequencies restrict the analysis in a near-surface region.
Applications of LT are numerous and in different fields. For example, interestingly, in addition
to photothermal heating, other heating employments are possible for various applications. In
case of medicine, analysis of blood circulation was demonstrated [78]. Here, a compression
cuff located in the upper arm creates the modulation, enabling us to analyze the blood flow in
the forearm: repetitive compression is performed through control of the air pressure going into
the cuff. Low frequencies (0.03 and 0.015 Hz) reveal blood vessel visualization down to 3 mm
under the skin. This has created interest in visualizing the functionality of the circulatory
50
system and especially to discover changes related to a developing pathological process.


2.3.4 Vibrothermography


The concept of vibrothermography (VT) originates from the German physicist W. E. Weber
(1804-1891), who discovered in 1830 that an increase in length of a material results in a
decrease in temperature. Such an idea was later formalized by the English physicists W. Kelvin
(1824-1907) in 1853 and J. Joule (1818- 1889) in 1857. Since this effect is reciprocal, a
reduction in length causes an increase in temperature. Periodical loading of the material thus
generates hot spots at locations of stress concentration.
Vibrothermography was first introduced as an application method in the late 1970’s.
Researchers observed that as cracks or discontinuities in a solid sample were excited with
high energy – low frequency ultrasound, they generated heat and were detectable with an
IR camera. Despite this promising start, the technique remained dormant until recently, when
techniques known as sonic thermography, or thermosonic testing were introduced. Significant
improvements in IR camera performance over the past 20 years have made it possible to detect
small cracks using lower excitation energies (or short duration pulses of high energy) than the
previous reports suggested. As a result, there has been considerable renewed interest in
implementing vibrothermography as a nondestructive crack detection method.
Vibrothermography (VT) is an active infrared imaging technique based on that principle: under
the effect of mechanical vibrations induced externally to a structure, thanks to direct
conversion from mechanical to thermal energy, heat is released by fraction precisely at
locations where defects such as cracks and delaminations are located, making them detectable
by an infrared camera.
In the basic scheme for modern vibrothermography-based testing, an acoustic horn is placed in
contact with a solid sample, and a brief pulse of acoustic energy in the 10 – 50 kHz range is
applied to the sample. Pulse energy is typically in the range of 500 – 3000 Joules, and duration
less than a second. The signal generation process in vibrothermography is almost entirely
determined by the interaction of the injected sonic energy with the mechanical properties of the
sample. The thermal and IR emission characteristics of the sample play a relatively small role.
In principle, heating of the sample is minimal unless a crack or some other source of frictional
51
heating is present. However, in practice, there is often significant localized heating at the sonic
insertion point, and also at any clamping points. Nevertheless, discrimination between cracks
and background features is relatively straightforward, compared to other conventional active
thermography methods.



Fig. 2.9 : Sketch of basic vibrothermography interaction. Acoustic energy from a horn is injected into
a solid sample, causing frictional heating at the tip, or along the faces of a crack.


In vibrothermography, flaws are excited at specific mechanical resonances. Local subplates
formed from delaminations resonate independent of the rest of the structure at particular
frequencies. Consequently, by changing (increasing or decreasing) the mechanical excitation
frequency, local thermal gradients may appear or disappear. Finite element modeling applying
3D equations of linear elasticity permits us to evaluate local energy concentration for
components submitted to mechanical loading.
However, in some circumstances, mechanical loading provokes breakdown of the sample.
This is explained by dynamic instability: one side of the hole takes more load than the other
and thus becomes hotter, while on the other side, loss of stress stop heat generation.
In vibrothermography technique most significant advantages are the detection of flaws hardly
visible by other infrared thermography schemes, and the ability to inspect large structural areas
in situ provided that required mechanical loading can be achieved.









52
2.4 Infrared detectors


An infrared detector or sensor is a device that converts radiant energy into some other
measurable form, e.g. an electrical current. There are two main classes of infrared sensors:
thermal and photonic (or quantum) detectors.
A thermal detector responds to incident radiation by raising its temperature. When a thermal
detector is at equilibrium there is no thermal conduction, the sensor simply radiates energy at
the same rate as it absorbs it. The detector’s temperature will increase whenever the incident
radiation rises above the equilibrium state, the detector absorbs more energy than it radiates.
The excess temperature is then removed away by conduction. Types of thermal detectors
include thermocouples, bolometers, pneumatic detectors, pyroelectric detectors and liquid
crystals. Thermal detectors response is theoretically independent of wavelengths and they
display high sensitivity at room temperature (see microbolometric spectral curve in fig. 2.11).
A photonic detector is made from semiconductor materials and operates on the principle that
incident radiation excites electrons from the valence to the conduction atomic bands. Even
though they provide superior sensitivity and fast response time than thermal sensors, photon
detectors typically require cryogenic cooling to minimize the noise to obtain the high relative
sensitivity.
In summary, thermal detectors exhibit a modest performance, with slower response than
photonic sensors, but they are less expensive and do not require cryogenic cooling as their
photonic counterparts. Developments in both fields predicts that uncooled photon detectors and
thermal detectors with better performance will soon emerge. In the next sections, firstly it will
be discussed the problem of the atmospheric trasmittance that define the accessible spectral
bands for infrared sensor and then, some thermal and photonic detectors of interest for infrared
imaging application are briefly presented.








53
2.4.1 Atmospheric Trasmittance


Atmosphere is a complex mix of various gases, particles, and aerosols in different
concentrations. These constituents both absorb and scatter radiation as it travels from the target
to the thermal imaging system. A description of the transmittance of electromagnetic energy
through this mix is thus a complicated task. The main problem concerns with the transmittance
of electromagnetic energy which is less than 100%. Self-absorption by gas in the atmosphere
(e.g . H
2
O, CO
2
, O
3
) and diffusion due to particles such as molecules and aerosols are
responsible for this. Moreover, various factors, such as the presence of thermal gradients and
turbolences, make the index of refraction inhomogeneous and contribute further to degradation
of the infrared measurement through the atmosphere.
Finally, the atmosphere itself emits infrared energy which is captured by the IR sensor. Most of
these effects depend on the distance, wavelength and meteorologic conditions, they are thus not
easily taken into account. In order of importance, main gases that absorb the radiation
transmitted are water vapor (H
2
O), then carbon dioxide (CO
2
) and ozone (O
3
).
Due to the atmospheric special transmittance, electronic imaging system design is partionated
into seven geneneric spectral regions. Four of these are associated with thermal imaging
systems. The ultraviolet (UV) region ranges in wavelength from 0.2 to 0.4 μm. The visible
region ranges in wavelength from 0.4 to 0.7 μm. Televisions, electronic cameras, and most
solid state cameras operate in this region. The near infrared imaging spectral region (NIR)
spans approximately from 0.7 to 1.1 μm. The first infrared imaging band is the short
wavelength band (SWIR) which approximately covers from 1.1 to 2.5 μm. The second infrared
band is the mid-wavelength infrared (MWIR), spectral region that covers approximately from
2.5 to 7.0 μm. The third infrared band is the long wavelength infrared (LWIR) spectral band. It
covers the spectral region from approximately from 7 to 15 μm. The fourth infrared band is the
far infrared or very long wave infrared (VLWIR) region. The last one applies to all systems
whose spectral response extends over 15 μm. The MWIR and LWIR regions are sometimes
called the first and second thermal imaging bands respectively.
Fig. 2.10 shows a typical atmospheric transmittance curve that accounts for absorption only.
The dominant absorber in the 3 to 5 μm region is carbon dioxide (CO
2
) which absorbs at 4.3
μm. This absorption band is obvious after a few meters of path length. Water vapor (H
2
O)
determines the upper and lower wavelength limits for both MWIR and LWIR spectral regions.
54
Water vapor is also the dominant absorber in the LWIR region. As the water vapor increases,
the LWIR transmittance decreases more quickly than the MWIR transmittance. This implies
that a MWIR system may be better in a tropical or maritime environment.


Fig. 2.10 : Atmospheric trasmittance and infrared imaging bands


The selection of the operating wavelength band is affected by the application. For the majority
of nondestructive-testing applications, the useful portion of the infrared spectrum lies in the
range from 0.8 to 15 μm, beyond 15 μm applications are more exotic, such as high-
performance Fourier trasform spectrometers, which operate around 25 μm. The choice of an
operating wavelength band is also important because dictates the selection of the detector type,
as fig. 2.11 shows.



Fig. 2.11: Spectral response curves for some infrared detectors in different infrared regions

SWIR


MWIR


LWIR
55
Among the important criteria for band selection there are operating distance, indoor-outdoor
operation, temperature, and emissivity of the bodies of interest. As Planck's law says,
high-temperature bodies emit more in the short wavelengths, consequently, long wavelengths
will be of more interest to observe near room-temperature objects. Emitted radiation from
ordinary objects at ambient temperature (300 K) peaks in this long-wavelength range. Long
wavelengths are also preferred for outdoor operation where signal are less affected by radiation
from the sun. For operation distances restricted to a few meters in the absence of fog or water
droplets, atmospheric absorption has little effect. Spectral emissivity is also of great
importance since it affects the emitted radiation.
Although no specific rule can be formulated, the most useful bands are from 3 to 5 μm (MWIR)
and from 8 to 12 μm (LWIR) since they match the atmospheric transmission band. Most of the
infrared commercial products and thermographic applications fall in these categories. The
SWIR band, instead, is principally used in applications where reflected infrared property of
samples are important.
However, detailed studies [79-81] have concluded that for temperature in the range
from -10 to +130 °C, measurements can be done without much difference in both bands (from
3 to 5 and from 8 to 12 μm). For some special applications (e.g. for the military), bispectral
cameras operating simultaneously in both bands have been developed to characterize target
thermal more accurately.



2.4.2 Bolometers

Bolometers are thermal detectors. For bolometers, such as thermistors, the relevant physical
property is the electrical conductivity. Consequently, a current must cross the element for
signal generation. It should, however, be noted that this current, having its own fluctuation,
adds to the detector noise. One of the main drawbacks of bolometers is in their slow response
time (from 1 to 100 ms), which restricts their use to slowly varying processes. The figure of
merit of bolometers is the resistive temperature coefficient defined as

dT
dR
R
B
B
1
= o
2.14

56
where R
B
is the internal resistence of the bolometers and T the temperature. The value of this
coefficient is either positive or negative, depending on the conduction mechanism taking place
within the material of interest. Typical values are 0.1% K
-1
for metal, from l to 10% K
-1
for
semiconductors, and higher values for superconductors. Recently arrays of microbolometers
have opened a whole range of new infrared imagers.




2.4.3 Pyroelectric detectors


Pyroelectricity is defined as the property of certain crystals to produce a state of electric
polarity by a change in temperature. A broad class of thermal detectors are pyroelectric
detectors. For these detectors below the Curie temperature
1
electric charges are generated by
incident radiation absorption (heating): a change in the detector temperature generates a
transient change in the surface charges, thus causing a transient current available for pickup by
the readout unit. Pyroelectric detectors are sensitive only to temperature variations and thus
need a chopper while modulates the incoming radiation in order to maintain a signal output,
unless the system is panned over the scene. This added complexity is a severe drawback. The
figure of merit of pyroelectric detectors is the pyroelectric coefficient p, defined as the slope of
the material polarization P versus temperature T at the operating temperature

dT
dP
p =
2.15


This effect can be improved with the application of an electrical field bias.






1
Temperature below which there is a spontaneous magnetization in the absence of an externally applied magnetic
field
57
2.4.4 Quantum Detectors

Quantum detectors are solid-state photonic detectors in which photon interactions either
change conductivity (photoconductor) or generate voltage (photoelectric, also called
photovoltaic). Sinœ no heating phenomenon is needed as in case of a thermal detector, the
response time is short while the solid-state structure made these detectors compact, reliable,
and robust. Consequently, they are quite popular. For photoconductive detectors an external
current is necessary to measure conductivity change while photoelectric (photovoltaic)
detectors act as a power generator, supplying a signal without need for polarization. Since
biasing current is needed for photoconductor detectors, high-charge-capacity alkaline batteries
can be used to induce minimum noise and ripple. This is essential to achieve stable results. In
this respect, since photovoltaic detectors supply a signal by themselves, they are much more
attractive than photoconductor detectors, requiring less complex readout circuits. Common
materials used in photoelectric (photodiodes and phototransistors) are InAs, InSb, GaAs, PtSi
and HgCdT. Fig. 2.11 presents the spectral response curves for the most common detectors.

- InSb
Indium antimonide (Insb) is a high quantum efficiency MWIR detector. It has generally
replaced PtSi in those systems using modest sized arrays (e.g. standard video format with
approximately 640x480 elements). It peak response is near 5 μm. A filter is used to limit the
spectral response to the MWIR region.

- HgCdTe
Mercury cadmium telluride (HgCdTe or MCT) is a mixture, in different ratio, of Hg
1-X
Cd
X
Te.
By varying the ratio, the spectral response can be tailored to the MWIR or LWIR region.
The most popular is the LWIR with a peak response near 12 μm. A filter is used to limit the
response to the LWIR region. HgCdTe detectors are used in all common module systems.

- QWIP
The quantum well infrared photodetectors (QWIP) is based upon GaAs growth technology.
The wells are created by layers of GaAs/AlGaAs and the response can be tailored from 3 to
19 μm. The LWIR version has a spectral responsivity from 8.3 to 10 μm. The responsivity and
noise are temperature sensitive so that QWIP devices are cooled to less than 60 K.
58
2.5 Infrared Imaging Devices: Focal Plane Arrays (FPA)

In 1980s a new kind of imaging device started to appear and revolutionize the infrared
community: large-dimensional infrared arrays simplified infrared camera construction.
Due yo their initial expensive cost, these devices have become popular only in the last fifteen
years. Focal Plane Arrays (FPAs) are today’s most common IR imaging devices configuration.
A FPA is made up of rows and columns of individual infrared detectors. Using this technology,
all what is needed to build an infrared camera is the optics, the focal plane array (FPA), the
associated electronics, and for some detector technologies, a cooling unit. Similar to
conventional video CCDs, these chips do not require any electromechanical scanning
mechanism (no moving parts) for image forming and are less fragile than pyroelectric tubes.
Video signals are obtained directly by on-chip electronics drive.
The fill factor is an important parameter in selecting FPAs, it provides the IR-sensitive material
to total surface (including the signal transmission paths) ratio. The higher the fill factor, the
higher the sensitivity, the cooling efficiency and the overall image quality of the system.
If FPAs have both IR-sensitive material and signal transmission paths on the same layer they
are called monolithic FPAs. Conversely, an hybrid array has the IR-sensitive detector material
on one layer and the signal-transmission and processing circuitry on another layer bonded
together by small collecting bumps. Hence, monolithic FPAs are easier and less expensive to
manufacture than hybrid FPAs, but they generally have lower performance since, having the
detector material and signal pathways on the same level significantly reduce the fill factor
(~55% for a monolithic FPA vs. ~90% for a hybrid FPA). The readout device is also important.
An FPA can use one of two technologies, a Charge Coupled Device (CCD) or a
Complementary Metal Oxide Semiconductor (CMOS). The principal advantages of CMOS
with respect to CCDs are: lower fabrication cost, possibility of developing a single-chip
camera, lower voltages and have lower power requirements (portability), possibility to read out
a portion of the array rather than the entire array (windowing). On the other hand, CMOS have
much higher noise than CCDs, since CMOS have more on-chip circuitry and uses amplifiers
for each pixel, producing nonuniformities known as Fixed Pattern Noise (FPN). FPN can be
removed by recording the pattern and post processing with software to remove it [rif], but this
further reduces the useful dynamic range. Furthermore, the sensitivity and the filling factor
(< 30%, the amplifiers also take up more area) of the CCD is much greater. CMOS is expected
to dominate in the low end consumer market (camcorders, snapshot cameras, toys, etc.) and
59
other specialty applications such as surveillance. However, CCDs will remain unchallenged for
scientific and technical applications requiring high fidelity, resolution, and dynamic range in
the next future.

- Microbolometer Arrays

ln the early 1980s, with the broader availability of large integrated-circuit fabrication
technology, it was possible to build large uncooled bidimensional arrays based on
microbolometers.
With micromilling technology, it is now possible to constitute arrays with thousands of tiny
bolometers called microbolometers, which are, in fact, thermal masses (also called
microbridges) hung by low-thermal-conductivity arms supported on metal legs anchored in the
silicon substrate, thus enabling a direct interface with readout electronics located under the
array (fig. 2.12). The IR radiation heats the hanging mass, modifying its electrical conductivity
so that a voltage or current variation appears at the device outputs. These arrays are, for
example, built using planar MOS or thin coating technologies adapted to micromilling. One
common technology is the VO
X
(vanadium oxide). Nowaday, ferroelectric microbolometer
arrays have performances such as time costant t on the order of a few milliseconds, detectivity
on the order of 10
7
to 10
9
cm Hz
1/2
W
-1
, NETD (Noise Equivalent Temperature Difference) 50
mK, and a pixel size of 25μm (since 1024x768 array size)








Fig. 2.12 : Microbolometers array





60
2.5.1 Detectors: Performance parameters

In order to compare different detectors available on the market, it is very useful to know the
most used figure of merit for infrared devices. A brief description of some of them is given in
this section.

- Impedence

The detector impedence Z is an intrinsic characteristic of detectors measured using Ohm’s law
dI
dV
Z =
2.16
This parameter is an important design property since the maximum power we can extract from
such a detector is reached when
A
Z Z =
2.17

Where Z
A
is the input impedance of the preamplifier to which the detector is connected. A
correct impedance match is important to maximize the output signal since electric signal at the
detector output are generally small.

- Responsivity

Another important parameter is the responsivity expressed in voltage (R
V
) or current (R
I
) and
which is the trasformation ratio of the incident optical flow F

dF
dV
R
V
=
dF
dI
R
I
=
2.18

Responsivity is generally not uniform on the detector area, moreover, it also depends on the
radiation wavelenght and excitation frequency.



61
c
f t
t
2
1
=
- Time costant

A useful property is also the time costant , which specifies the response time of a detector.
It is given by
2.19

f
c
is the critical frequency defined by an observed decrease of 3 dB in the maximum amplitude.

- Noise Equivalent Power

A very important figure of merit for detectors is the noise equivalent power (NEP). This the
amount of power induced by an optical signal whose amplitude is equivalent to intrinsic noise
power present at the detector output without signal, which originates from thermal agitation
and optical radiation granularity.

- Noise Equivalent Temperature Difference

Many detector manufacturers rate their detectors using the noise equivalent temperature
difference (NETD) instead of NEP. NETD correspond to the change in temperature of a large
blackbody
2
in the observed scene, causing a change in the signal-to-noise-ratio of unity in the
output of the detector. NETD depends on the F-number
3
of the optics and on the pixel size.

- Detectivity

The detectivity D is also a widely used parameter to specify detectors. It is given by
NEP
F
D
1
=
2.20
When F
NEP
is the optical flow called the noise equivalent flow. D should be high, to obtain a
reasonable signal-to-noise ratio. The detectivity depends on many parameters : spectral content
and modulation of the incident radiation, receiver system bandwidth, detector temperature, and
sensitive surface area.

2
Perfect emitter of infrared radiation with max value of emissivity
3
The F-number of a lens is given by its focal distance divided by its effective diameter aperture
62
2.6 New approach of infrared imaging in medical science

Infrared imaging allows the representation of the surface thermal distribution of the human
body. Several studies have been performed so far to assess the contribution that this kind of
information may give to clinicians. During the last 30 years it has been widely discussed
whether or not infrared imaging should be considered a good diagnostic tool. The approach
generally followed consisted of the detection of significant differences between the skin
thermal distributions of the two hemisomus or in the evaluation of the mismatches of the skin
thermal distribution with respect to an average extrapolated from a control group. So, after the
contradictory results obtained during the 1970s, a general lack of confidence has arisen among
clinicians regarding the clinical use of infrared imaging.
Simultaneously, but for other purposes, technological improvements released a new generation
of digital infrared cameras characterized by :

- high spatial resolution
- optimized temperature sensitivity
- very short acquisition time

By means of the last generation infrared cameras, it is possible to record in real time all the
modifications of the skin thermal distribution following a physiological or environmental
processes; i.e. the action of vasodilator or vasoconstrictor substances or the execution of
thermal stresses.
A dynamic approach to infrared imaging by means of the characterization of the dynamics of
the thermal recovery is therefore possible [82]. This permits to overcome the limitation of
conventional infrared imaging and to use a new approch based on the modellization of
functional parameters [83].
The functional infrared imaging is the study of the functional properties and alterations of the
human thermoregulatory system based on the modelling of the bio-heat exchange processes.
The skin temperature distribution of the human body depends on the complex relationships
defining the heat exchange processes between skin tissues, inner tissues, local vasculature, and
metabolic activity. All these processes are mediated and regulated by the sympathetic and
parasympathetic activity to maintain the thermal homeostasis. The presence of a disease can
locally affect the heat balance or exchange processes resulting in an increase or in a decrease of
63
the skin temperature. Such a temperature change can be better estimated with respect to the
surrounding regions or the unaffected contra lateral region. But then, the dynamics of the local
control of the skin temperature should be also influenced by the presence of the disease.
Therefore, the characteristic parameters modelling the activity of the skin thermoregulatory
system can be used as diagnostic parameters.
The aims of this new approach of infrared imaging is highlighting the functional contents of
skin thermal distribution, by evaluating qualitatevely and quantitatively how the lesion or the
disease alters the thermoregulatory response of the damaged or ill area with respect to the
healthy ones.
Several studies have been performed so far to assess the contribution that this new approach
may provide to the medical science. Infrared technology is used for:

- Valuation of re-perfusion process during surgery (thermo-angiography) [84]
- Follow up of the modification of the neurovascular functions of the treated district by
means of the study of thermoregulatory properties [85, 86]
- Assistance to cardiac surgery and microsurgery, for wich the real time evaluation of the
re-perfusion processes and the estimate of the ricovery of the neurovascular functions
constitute important information during the follow up [84, 87]
- Validation and detection of sub-clinical varicocele [88-90]

Moreover, Infrared Imaging was used to record the temperature variation of the hands
associated with sympathetic skin response [91]. The submotor thermal response may he used in
the future as complementary or alternative monitoring of the activity of the autonomous system
in conditions where the traditional sympathetic skin response cannot be recorded.
The new infrared approach is used, also, to determine the major infrared thermographic
parameters in discriminating between patients with and without secondary Raynaud’s
phenomenon. Some of these regard the Raynaud’s phenomenon secondary to Scleroderma
(SSc) or chronic vasospastic syndromes [92-95].
Recently, a novel method of estimation of blood flow speed and vessel location from thermal
imaging recording has been developed. The method is based on a bioheat transfer model that
reflects the thermo-physiological processes in a skin region proximal to a major vessel [96].
A new possibility on the remote control of the humans functional parameters is given from a
novel method to measure human cardiac pulse at a certain distance. It is based on the
64
information contained in the thermal signal emitted from major superficial vessels [97-99]. The
contribution that the functional infrared imaging can give in the diagnosis of the breast cancer
has not been clarified. Some studies show how as the infrared imaging, thanks to pattern
recognition methods, can individualize the tumoral neoangiogenesis and the excessive
production from endotelium of nitric oxide associated to the lesion. Anyway, the capacity to
localize the tumor with this technique is subordinates to the realization of new realistic model
of the thermal diffusion in vivo [100-102].
The technology is expected to find applications among others in sustained physiological
monitoring of cardiopulmonary diseases, sport training, sleep studies, and psychophysiology
(polygraph).
All these studies (and many others) has proved that infrared thecnology can be advantageously
used in these field thanks to its new features, overall thanks to new dinamyc approach and new
images elaboration methods.




















65
Chapter 3

INFRARED MONITORING ON PLANE ANGIOMA :
PT model and numerical simulation

3.1 Introduction

In the first chapters the problems concerning to laser applications in medical fields and infrared
imaging technologies have been described.
In this chapter, instead, it will be given a detailed description of the activity research, object of
the present thesis-work, in which the laser - biological tissue interaction on patients affected by
plane angioma pathology has been monitored and studied with the infrared imaging technology.
Subsequently, the approach employed, the pathology analized and computation instruments
used (pulsed thermography model and a numerical simulation) will be treated and discussed.
This activity has been realized in collaboration with the LIRT of the Cybernetic Institute “E.
Caianiello” - ICIB-CNR and in collaboration with the dermatological division of the II
Policlinic in Naples in the laser therapy and photodynamics ambulatory.



3.2 Infrared monitoring of plane angioma treatment: approach description

In the last years, the employment of differents laser systems for more and more medical
treatments has had an esponential growth. The effects of the laser radiation on differents
organics tissues aren’t easily standardizable, because of dermatological differences showed
from differents anatomics regions of the body and in differents people. This indetermination
influences the therapeutics results obtained, and, in the majority of cases, they are
unsatisfactories or even harmfuls. The scientific research in this field is addressed to obtain
quantitative and functionals informations about the dynamic of this interaction in order to
allow to specialists to employ laser systems in a more specific, well controlled and less empiric
way.
66
Infrared imaging technology can be used, in this field, for a real-time monitoring of the
laser-biological tissue interaction. This technique can give to specialist information about the
reaction of the tissue in the specific region treated.
In this activity the main pathology monitored and studied with infrared technique has been the
Plane Angioma known as Port Wine Stain (PWS) pathology. This pathology, as described in
the next section, is a vascular malformation that consists in a blood vessels accumulation under
the tissue. This accumulation forms a subsurface plane of vessels of depth and thickness
depending on the specific anatomic region and, in general, differents from case to case.
The main problems in the cure of plane angioma with laser therapy are:

- The therapeutic efficiency can differ a lot from patient to patient. In general it can
depend on the age, anatomic region, phototype, lesion depth and lesion thickness.
Usually, in the median region of the face the therapeutic efficiency is better then
external area.

- Possible formation of scars

- Possible formation of dischromic area

In the laser treatments on patients affected by Plane Angioma the therapeutic program and laser
parameters are chosen following general tables and indications present in literature. In the next
table it is shown the reference range for laser parameters advised in literature for the treatment
of patients with Plane Angioma:










Tab. 3.1: Laser parameters for Plane Angioma treatments

Referential range for laser parameters

Wavelength

585 – 595 nm

Energy density

4 – 10 J/cm2

Pulse duration

1 – 5 ms

Spot diameter

2 - 7 mm
67
Even if modern laser system permitted more and more control on the operative variables,
nowadays it is the operator who chooses the system and the operative protocol. In general, the
specialist starts from the laser parameters showed in the table and then he changes these
parameters based on his own experiences and so in a very empiric way.

In the research activity, object of this work-thesis, the effects of the laser-biological tissue
interaction on PWS patients have been studied through the information obtained with infrared
imaging technique.
With this technique it’s possible to obtain information about a specific plane of vessels under
treatment monitoring the thermal reaction of the tissue under laser action. The main purposes
of this study have been:

- To evaluate the support of the infrared imaging, in the dermtological field, as an
efficient tool for scientific investigation

And particularly

- To evaluate the employment of this technique as a tool for the control and
optimization of laser parameters in order to make the treatment more safety and
more efficient

So, with the intent of optimazed laser parameters for each specific plane of vessels treated by
specialist, the next approach has been studied.
First of all, in order to find morphological information for the specific plane of vessels under
treatment a Pulsed Thermography model has been used.
With this model, in fact, from the temporal evolution of the surface’s temperature obtained
after a single laser pulse-test (heat souce), it is possible to get a local evaluation of the depth
and thickness of the plane angioma in the anatomic area that the physician wants to treat. This
information permits a morphological reconstruction of the multilayer biological tissue under
treatment.
Subsequently, it is executed on this multilayer a 2D+1 numerical simulation based on a
biometric heat transfer model. In the simulations the thermal behaviour of the biological
multilayers can be achieved with different set of laser parameters as input.
68

The numerical output consists of the evaluations of the spatials and temporal temperature
profiles.
Finally, these results can be used:

- To compare the thermal behaviour simulated on the multilayer’s surface with the
experimental data (recorded with a infrared camera) in order to check the
approach performed.

- To optimize the laser parameters in order to achieve suitable local temperatures in the
multilayer for the photocoagulation of the blood vessels and for reducing the risk of scars
formation

Schematically, this approach can be synthesized in the next operative flow-chart :































- Single Pulse test on the plane angioma region under
treatment with the reference set of laser parameters

- Recording of the temporal evolution with infrared camera
{
{
{
- Evaluation of the plane angioma depth (d) and thickness (h)
by PT model, from the temporal evolution recorded

- Reconstruction of the biological multilayer under treatment

- Numerical simulation on the biological multilayer
starting with the referential set of laser parameters

- Comparision of simulated and experimental results
- Local temperatures evaluation

- Variation of the laser parameters in order to optimize
the local temperatures obtained
{
69
As it is possible to see from the flow-chart, in this operative approach, the results achieved
from simulations are used as a feedback in a loop in order to optimize laser parameters and so
to optimize the therapeutic efficency.
In the next sections the plane angioma pathology, the pulsed thermography model and the
numerical simulation used in this approach are described and discussed extensively.






















70
3.3 Plane angioma pathology (Port Wine Stain)

In the last 20
th
years, one of the field in which is much more
concentrated the dermatological interest has been the treatment of the
skin vascular lesion as plane angiomas (knowen as Port Wine Stain or
Nevus Flammeus too). As it is well known, the plane angioma is a
benign tumor made of blood vessels accumulation newformed and
malformed. This vascular malformation, is congenital and it is present
in about 3 newborns on 1000 with the same incidence for male and
female. PWSs are present at birth and persist throughout life. Rarely,
acquired PWS can occur at any age after birth and it is identical to
congenital capillary malformations both clinically and histologically.
The exact mechanism (causes) remains unknown. Evidence for genetic influence is lacking.
Capillary malformations may result from a neural deficiency of sympathetic innervation of the
superficial dermal blood vessels. The mean dimensions of the blood vessels in a plane angiona
are reported in the next table


Blood vessels dimensions

Thickness

20 – 180 μm

Depth

200 – 1300 μm

Tab. 3.2 : Mean dimensions of the blood vessels in a plane angioma

The area of the skin affected grows in size commensurate with the general growth. Plane
angioma occur most often on the face but can appear anywhere on the body. Early stains are
usually flat and pink in appearance. As the child matures, the color may deepen to a dark red or
purplish colour. The color of the capillary malformation does not correlate with the capillary
depth or diameter. In adulthood, thickening of the lesion or the development of small lumps
71
may occur. A physician can usually diagnose a PWS based entirely upon the history and
appearance. In unusual cases, a skin biopsy may be needed to confirm the diagnosis.
Many treatments have been tried for PWS including freezing, surgery, radiation, and tattooing,
(plane angioma can also be covered with cosmetics). Lasers have had the biggest impact on
treatment, because they are the unique method for destroying the cutaneous capillaries without
a significant damage to the overlying skin. During the years many type of laser system have
been used: CO
2 ,
Argon laser, Nd:Yag and others, but several study and experiences have
highlighted as the FPDL (Flash-Pump-Dye-Laser) is, at present, the laser system tool more
efficient and less dangerous for curing PWS patients. Treatment of infants with the FPDL
generally produces marked improvement in appearance. However, in general, complete
disappearance is rare.
Usually, in approximately 20-25 % of cases there is an optimal improvement, in 30-35 % the
improvement is acceptable and, unfortunately, in approximately 50%

of cases the therapy is
less incisive and there may be no improvement. Stains on the face respond better than those on
the trunk or limbs. Older stains may be more difficult to treat. In the absence of successful
treatment, hypertrophy (increased tissue mass) of the stains may occasionally produce
deformity, loss of function (especially near the eye or mouth), bleeding, and increasing
disfigurement. These complications are usually seen later in life and psychosocial disability
secondary to facial disfigurement can be overwhelming. If the PWS is on the face or other
highly visible part of the body, the presence of PWS cause emotional and social problems for
the affected person because of their cosmetic appearance. Several studies demonstrate that
patients with facial capillary malformations exhibit greater self-concern, ruminative self-doubt
in interpersonal interactions, social inhibition, isolated and passive orientation in interpersonal
relationships, stigmatization from society, and limitations of privileges and opportunities
otherwise afforded to those without facial disfigurement. A study demonstrated that the
psychosocial difficulties not only persisted but actually worsened in adulthood [103].
In same cases (about 5%) plane angiomas can be a cutaneous finding of several syndromes as:
Sturge-Weber syndrome, Klippel-Trenaunay syndrome, Proteus syndrome, Parkes-Weber
syndrome, Von Hippel – Lindau syndrome, Cobb sindrome, Wyburn-Mason sindrome and
others.



72
3.4 Pulsed Thermography model

In this work, in order to find unknown parameters for each specific subsurface plane of vessels,
it has been used a tested modelling approach of pulsed thermography (PT) based on the use of
the 3D thermal quadrupole formalism [56, 57]. In general, with this approach, from the
temporal evolution of the superficial temperature, it is possible to estimate the depth and the
thickness of a subsurface anomaly or disomogeneous area present in a structure. In the
multilayer biological structures analyzed in this work, the role of subsurface anomaly is
rappresented by the plane of vassels associated with the presence of a plane angioma [rif ben].
In this model, three-dimensional thermal quadrupoles are obtained by applying to the real
temperature field a Laplace transform in the time domain, then a double Fourier transform in
the space domain is applied to the previous Laplace temperature field. After the last three
integral transforms, the temperature field is treated in the Laplace–Fourier–Fourier image
domain.
A plane of blood vessels, in a multi-layer biological tissue, is thermally characterized by a
thermal contact resistance. Because this resistance is small compared with the tissue whole
resistance, a perturbation method [104-106] can be applied and combined with the thermal
quadrupole formalism to yield simple analytical solutions of the nondestructive experiment.
The perturbation procedure is based on an asymptotic expansion of the physical field
(temperature or heat flux) versus a small parameter intervening in the model. The analytical
solutions obtained via this method are very convenient for the development of new inversion
procedures which characterize the subsurface anomalies within a three-dimensional heat
transfer configuration.
Resuming, in this model, in order to accurately identify the unknow parameters associated to
anomaly, two stages have been considered:

1) The first step is to develop a forward model which consists of describing the contrast field
evolution in a mathematical formalism as accurately, simply and rapidly as possible

2) The second step is to develop an inverse problem that estimates the unknown parameters in
order to minimize the deviation between the experimental data and the forward problem

These two important steps will be accurately described in the next subsections.
73
3.4.1 Formulation of the forward problem: 3D thermal quadrupoles

This model refer to the case of a rectangular (L
1
× L
2
) multilayer sample of thickness d
1
that
contains a resistive anomaly of finite width a and finite length b, with a uniform contact
resistance R
c
on its whole area, typical of biological tissue affected by plane angioma.


Fig. 3.1: Multilayer sample of thickness d
1
, with a plane angioma depth d and thick h
v

In a typical pulsed infrared thermography test, one can assume that:

1 - the thermal excitation of the sample is a Dirac distribution characterized by a uniform
absorbed energy Q (at time t = 0)
2 - the front face of the sample is adiabatic
3 - the sample temperature is zero before excitation

Since only the face temperatures are of interest, the quadrupole technique is the easiest
formalism to implement in order to describe the heat transfer. Thermal quadrupoles are
obtained after performing a Laplace integral transform on the time variable t, and a double
Fourier integral transform on the space variables x and y [23, 24 riferim]. Heat transfer
modelling of multi-layer materials is then reduced to a simple multiplication of matrices in the
transformed Laplace–Fourier–Fourier domain. First of all, recall here the heat equation in 2+1
dimensions:


3.1


0
2
2
2
2
2
2
=
c
c
÷
c
c
+
c
c
+
c
c
t
T
y
T
k
x
T
k
z
T
k
y x z
74
where T is the temperature and k is the thermal conductivity. If the initial temperature is
uniform and if the heat losses are neglected, the Laplace transform τ(x, y, z, p) (where p is the
Laplace variable) of the temperature T (x, y, z, t) in the sample is the solution of the following
set of equations:





3.2






with s(x, y) = 1 for (x, y) inside the anomaly region [(x
1,
x
2
) , (y
1,
y
2
)] and s(x, y) = 0
elsewhere, d is the depth of the anomaly. Superscripts sup and inf relate to the upper and lower
face of the anomaly


3.3

The analysis can simultaneously be performed for both isotropic and anisotropic materials
(providing that the sample faces are parallel to the main directions of anisotropy) by using the
following dimensionless governing parameters:



3.4



( )

c
c
÷ = ÷
c
c
=
c
c
÷ =
=
c
c
÷ ÷ =
=
c
c
÷ =
=
c
c
÷ =
= ÷
c
c
+
c
c
+
c
c
z
k y x s R
z z
d z
Q
z
k z
y
L y
x
L x
k
p
y
k
k
x
k
k
z
z c
z
z z
y
z
x
t
t t
t t
t
t
t
t
t t t
,
0
0 , 0
0 , 0
0
inf sup
inf sup
2
1
2
2
2
2
2
2
0
1
=
c
c
÷ =
z
d z
t
( )
( )
( )
1
*
2 1
1
*
2 1
1
*
*
1
*
d z z
k k
d
y
y
k k
d
x
x
Q
k Qd
y z
x z
z
=
=
=
=
=
v v
t t
( )
( )
( )
1
*
1
*
2
1
*
2 1
1
*
2 1
1
*
d d d
k d R R
k p d p
k k
d
b
b
k k
d
a
a
z c c
z
y z
x z
=
=
=
=
=
75
where ψ is the Laplace transform of the z component of the heat flux density =-k
z
( T/ z).
In the remaining text of this chapter, the asterisk superscript in the above parameters will be
omitted for clarity reasons. The partial differential equation in 3.2 then becomes

3.5


The next step is to apply a Fourier transform to the function τ (x, y, z, p). The lateral boundary
conditions determine the type of transform to carry out [25], in this case it is a cosine shape

3.6


As for the boundary conditions on the opposite faces x = L
1
and y =L
2
, they determine the
discrete values allowed for the space frequencies: α
J
= jπ/L
1
, β
k
= kπ/L
2
, j and k are
non-negative integers.
Using equation (3.6) and the lateral boundary conditions, equation (3.5) yields

3.7


The solution of equation 3.7 has the following form:

3.8


where u =( p + α
2
+ β
2
)
1/2
and F and G are two constants that can be determined using the
boundary and interface conditions in z. If θ is known, back inversion into the Laplace domain
is given by

3.9

where θ
jk
= θ(α
j
, β
k
, z, p).
0
2
2
2
2
2
2
= ÷
c
c
+
c
c
+
c
c
t
t t t
p
y x z
( ) ( ) ( )dxdy y x p z y x p z q
L L
| o t | o cos cos , , , ) , , , (
1
0
2
0
í í
=
( ) 0
2 2
2
2
= + + ÷ u | o
u
p
dz
d
( ) ( ) uz G uz F sinh cosh + = u
( ) ( ) ( )

¯ ¯ ¯ + + ¯ + =
·
=
·
=
·
=
·
= 1 1 1
0
1
0 00
2 1
cos cos 4 ) cos( 2 ) cos( 2
1
, , ,
k j k
k j jk k k
j
j j
y x y x
L L
p z y x | o u | u o u u t
76
If is the Fourier transform of the Laplace heat flux density ψ, which is unity at z = 0 and zero
at z = 1 (equations 3.2 and 3.3) we obtain


3.10


If the arguments other than z are omitted in the Laplace–Fourier transforms θ and , this
equation leads to a linear relationship between the two quantities on the front (z = 0) and rear
(z = 1) faces


3.11

Boundary conditions in (3.2) and (3.3) can be rewritten under the following matrix forms,
commonly called quadrupoles [107]:



3.12






where



where d
1
and d
2
are the thicknesses of the two layers of the sample.
A simpler and faster method to obtain a solution of the equations (3.12) is the perturbation
method. In the following section, this approach is presented and analyzed.
( )
( ) ( )
( ) 0 , 1 , , 1
sin sin
, 0 , , 0
2 1
= ÷ =
= ÷ =
p z
L L
p z
| o o
|
|
o
o
| o o
( ) ( ) ( )
( ) ( ) ( ) 1 1 0
1 1 0
o u o
o u u
D C
B A
+ =
+ =
( )
( ) ( )
( )
( ) ( )
( )
( )

=

+
=

=

0
1
sin sin
0
2 2
2 2
inf
sup sup
sup
1 1
1 1
2 1
u
o
u
o
u
o
u
o
u
|
|
o
o
u
D C
B A
d
d
I R
d
d
D C
B A
L L
c
( ) ( ) ( )dxdy y x p d y x I
x
x
y
y
| o v cos cos , , ,
2
1
2
1
í í
=
( ) ( ) ( ) 2 , 1 sinh sinh
1
cosh = = = = = i for ud u C ud
u
B ud D A
i i i i i i i
77
3.4.2 Resolution of the forward problem: mathematical perturbations

The perturbation method consists of writing asymptotic series expansions of the variables θ
and with respect to a small parameter in the model. These series expansions are then injected
into set (3.12), and term-by-term identification of the coefficients of the successive powers of
the small parameter leads to a coupled series of linear sets. These allow the calculation of the
different components θ
i
and
i
. The mathematical formalism presented below is valid only for
small values of the thermal contact resistance R
c
. The asymptotic series expansions with
respect to the small parameter R
c
that will be noted ε from now on are


3.13


Term-by-term identification of the coefficients of ε
n
leads to the following quadrupole
equations:

For ε
0
-order identification set (3.11) results in


3.14




For ε1-order identification set (3.12) turns out to be




3.15


( ) ( )
( ) ( ) ¯ =
¯ =
·
=
·
=
0
0
, , , , , ,
, , , , , ,
i
i
i
i
i
i
p z p z
p z p z
c | o o | o o
c | o u | o u
( )
( ) ( )
( )

=

0
1
sin sin
0
0
2 2
2 2
1 1
1 1
2 1
u
|
|
o
o
u
D C
B A
D C
B A
L L
( )
( )
( ) ( )
( )
( )

=

+
=

=

0
1
0
0
1
2 2
2 2
1
inf
1
1
0
inf
1
1
sup
1
1
sup
1 1
1 1 1
u
o
u
o
u
o
u
o
u
u
D C
B A
d
d
I
d
d
D C
B A
( ) ( ) ( ) p e p e sinh / sinh
2 1 0
= v
78
I
0
has the same definition as I in equation (3.12), just replacing ψ by ψ
0
. The substitution of ψ
0
defined by equation (3.14) into this definition allows the calculation of integral I
0
and therefore
the solution of set (3.15).
The first-order temperature on the front face is given by:



3.16

where



3.17


Product of equation 3.16 by ε represent the Laplace–Fourier transforms θ = εθ
1
of the contrast
T(x, y, z
s
, t) on the front face. Return to the original (x, y, z
s
, t) domain can be achieved
numerically by using fast Fourier-transform (FFT) and Stehfest algorithms [108].
Let us remind that the modelling described previously is based on the assumption of an
adiabatic heat transfer. In reality, nondestructive experiments are always affected by heat losses.
However, as reported in the literature [109], the losses effect on thermal contrasts are very
small. This last result is a consequence of the nature of the thermal contrast itself: even if each
surface temperature T and T
0
might be strongly affected by heat losses, their difference,
T = T − T
0
, leads in general to a compensation for this effect.
Numerical simulations have been carried out in a previous work [106] to assess the
effectiveness of the above model. The simulations have shown that the perturbed contrast
profiles are in excellent agreement with the true ones for small thermal resistance values, with
ε 0.3. However, for higher resistances (ε > 0.3) the approximation is no longer appropriate,
confirming the restriction of the above model to weak anomalies or defects.
We can observe, that in the case of anomaly rappresented by a plane of vessels, characteristic
of a plane angioma, this condition is accomplished.
( )
( ) ( ) u p
d p ud
K
sinh sinh
) sinh( ) sinh( 4
0
2 2
o|
u =
|
.
|

\
| +
|
.
|

\
| ÷
|
.
|

\
| +
|
.
|

\
| ÷
=
2
cos
2
sin
2
cos
2
sin
1 2 1 2 1 2 1 2
y y y y x x x x
K | | o o
79
In fact, in the case of a plane of vessels thick h
v
=200 μm, inside a tissue thick h
t
1.5 mm
(maximum depth of laser interaction in a biological tissue for a wavelength =595 nm) with a
thermal conductivity for the plane of vessels and for the tissue respectively of k
v
=0.492 10
-8
W
m
-1
K
-1
, k
t
=0.322 10
-8
W m
-1
K
-1
, the normalized thermal resistance ε is given by



3.18


In the following sections, it is described the inversion procedures that allow the evaluation of
the depth and the thickness of the subsurface plane of vessels. These parameters are directly
derived from the perturbation 3D modelling described in this section.





















3 . 0 087 . 0 < = =
v
t
t
v
k
k
h
h
c
80
3.4.3 Evaluation of the depth

In this section, it is shown how the depth of the anomaly can be derived from the space
averaging of the Laplace contrast profile τ(x, y,zs , p).
The space average of the Laplace contrast is defined as


3.19

where τ is the difference between the Laplace transform of the temperature on one spot of the
face of the sample and its value in the absence of anomalies. With the previous firstorder
perturbation method, the integral in the right-hand side of equation (3.19) is equal to εθ
1
(0, 0, z
s
,
p). In the case of expansive plane of vassels ab/L
1
L
2
=1. The application of equation (3.16)
allows the calculation of τ for the front face:



3.20

Equation (3.20) written for two values p
1
and p
2
=4p
1
yields the analytical elimination of the
unknown parameters ε and. If m
1
and m
2
are the experimental average Laplace contrasts on the
front face, the depth d of the flaw is the solution of the following equation:


3.21
where


This equation leads to the following estimation of the anomaly depth:

3.22

( )
( ) ( )
( ) p
d p
p
2
2
sinh
1 sinh
, 0
÷
= A c t
( ) ( ) ( )
1
2 1
1
2
1
cosh 1 cosh p
m
m
d p
|
|
.
|

\
|
= ÷
( ) 2 , 1 , 0 , , = A = i and p y x m
i i
t
( ) ( ) ] 1 cosh cosh ln[
1
1
2 1
1
2
1
2
1
2 1
1
2
1
|
|
.
|

\
|
÷ +
|
|
.
|

\
|
÷ = p
m
m
p
m
m
p
d
dxdy p z y x
L L
s
í í
A = A
1
0
2
0
) , , , ( t t
81
3.4.4 Evaluation of the thickness

In this section, it is shown how the thickness of the anomaly can be derived from the space
averaging of the Laplace contrast profile τ(x, y, z
s
, p), if the depth of the anomaly d has been
already calculated. The thermal contact resistance per surface unit is defined as the ratio of the
anomaly thickness h
v
and its thermal conductivity k
v
, ε =h
v
/ k
v
. In case of anomaly
rappresented by plane of blood vassels, k
v
is simply the thermal conductivity of the blood,
which is a known parameter. Evaluation of the thickness h
v
is therefore equivalent to the
estimation of its thermal resistance, ε.
The equation (3.20) shows that the average Laplace contrast (and therefore the instantaneous
averaged contrast τ ) is proportional to the thermal resistance ε of the flaw. In other words, the
average Laplace contrast is proportional to the thickness of the plane of vessels.
Equation 3.20 depends on two unknowns: the thermal resistance ε, and the depth d. If it is
written for two values p
1
and p
2
= 4p
1
of the Laplace variable, the parameter d can then be
eliminated from the two corresponding equations by using the duplication properties of the
hyperbolic functions. If m
1
and m
2
are the experimental Laplace contrasts on the front face at
the centre of the flaw, the thermal resistance of the plane of vessels is given by the following
equation:


3.23


where









( ) ( )
( )
1 1 2
1 1
2 2
1
cosh
tanh sinh
m p m
p p m
÷
= c
( ) 2 , 1 , 0 , , = A = i and p y x m
i c c i
t
82
3.4.5 Statistical analysis of the accuracy of the inversion procedures

Any inversion algorithm produces estimated parameters. A measurement of the quality of the
inversion requires a quantification of the estimation error that has been made. This error can be
divided into two parts: the error caused by measurement noise, and the error caused by
performing the quadratures in time to calculate the Laplace transform and in space to calculate
the average Laplace contrast, on a limited number of points. In the following, this second error
will be neglected since the numbers of points, in time and space, used to perform the
quadratures are considered large enough. Using a stochastic analysis with the assumptions that
the noise on the thermal contrast T is additive, uncorrelated and of constant standard
deviation σ, it is possible to quantify the inversion errors caused by measurement noise. The
effects of measurement noise on the precision of the thermal resistance and depth are
represented by the variance equations 3.24 and 3.25




3.24






3.25


In figure 3.2, the relative standard deviation, σ /ε, σ
d
/d, of the thermal resistance and of the
depth are reported as a function of the Laplace variable p (= p
1
) for different values of the
anomaly depth. In the figure 3.2 the value d
1
refers to maximum length of laser interaction in
the tissue ( 1.5 mm for a laser wavelength =595 nm).

With regard to the other parameters in
equations 3.24 and 3.25, we used a standard deviation of the normalized T contrast σ = 0.1,
( )
|
|
.
|

\
|
|
|
|
|
.
|

\
|
+
+
A =
1
1
1 2 1
2 1 1
2 2 2
2
1 1
1
2
1
V
U
p p p
p p p
UV t o c o
c
( )
( ) | | ( ) | |
( ) | | ( )
( ) | | ( ) | |
1 2 1 2 1 2
1 1 1 1
1
1 2 1 1 1 1 1
1 2 1
1
cosh sinh 1
cosh sinh 1
cosh sinh 1
cosh 2
p m m p p m
p p p m
V
p m m p p m m
p m m
U
+ ÷ +
+
=
+ ÷ +
+ ÷
=
( ) ( )
( ) ( )
( ) ( ) ( ) ( )
( ) ( )
|
|
.
|

\
|
¦
)
¦
`
¹
¦
¹
¦
´
¦
÷ ÷
|
|
.
|

\
|
÷ ÷ ÷
÷
+ ÷
A
=
2
2
2
1
1
1
2
2
2
1 exp
1 cosh 1 exp
1 cosh
1 1 cosh
1
4
V
U
d p
d p d p
d p
d p
p
NM
t
d
c
o
o
( )
( ) ( )
( )
( ) ( )
( ) ( )
( ) ( ) ( ) ( ) ( ) d p d p p p
p p
V
d p p
p
d p p
p
U
÷ ÷ +
÷ =
÷
+
÷
=
1 sinh 1 sinh
sinh sinh 2
1 sinh 2
sinh
1 sinh 2
sinh
2
2
1
2
2 1
2
2
1
2
2
2
4
2
2
4
1
4
1
1
4
2
83
a normalized time step t = 4 × 10
-3
that corresponds a scan rate of 40 ms, a normalized
thermal resistance ε = 0.087, the numbers M and N are both 256.

a) b)










Fig. 3.2: relative standard deviation as a function of the Laplace variable p (= p
1
) for different
values of the anomaly depth. a) σ
d
/d for the depth, b) σ /ε for the thermal resistance

We can note from the figures that:

From fig. 3.2.a
1) The choice of the optimum p-value depends on the depth d, but it is always in the range 1–2
for deep anomalies.
2) A bad choice of p leads only to a small error on the precision for plane of vessels located
near the heated surface, whereas for deeper plane of vessels, the discrepancies are higher.
3) For the deep anomaly the error can changes from 3% for d = 0.1d
1
up to 20% for d = 0.7d
1.

4) the estimation error changes nonlinearly with the Laplace variable p and the flaw depth d.

From fig. 3.2.b
1) The optimum p-value that should be used for thermal resistance inversion independently
from the anomaly depth must be within the range 1–2.
2) The estimation of the thermal resistance is better for more superficial plane of vessels, for
optimum p-values the error changes from 25% for d = 0.1d
1
up to 75% for d = 0.7d
1
.

84
3.5 Numerical simulation: biometric heat transfer model

The morphological informations, obtained for different plane of vessels with PT approach,
have been used as input data for a numerical simulation based on a biometric heat transfer
model. This model describes heat transfer in the vicinity of a large plane of vessels proximal to
the skin. The mathematical problem is to retrieve the unknown spatial and temporal
temperature profiles associated with a specific multi-layer biological structure and for a given
set of laser parameters. It is assumed that the vessels act as a volumetric heat source for the
surrounding four-layer tissue structure. These layers are successively, in positive z direction,
the epidermis, the dermis, the plane of vessels, and the dermis again (core).


Fig. 3.3: Multilayer biological tissue

It has been also assumed that each layer is isotropic with respect to thermal conductivity k(z),
metabolic heat rate q
M
, density ρ, and specific heat c of the tissue. The heat effect of the vessels
on the skin temperature depends on the vessel’s location (depth) and from its thickness as well
as the blood flow speed and temperature. The plane of vessels can be considered a single large
vessel running along the x direction without traversing across the z direction (see Figure 3.3).
The thermal conduction in the tissue surrounding the plane of vessel is dominant in directions
parallel (x) and perpendicular (z) to the skin. We can neglect heat transfer along the (y) axis
because of the presence of other vessels, periodically arranged and similar to that considered.
Therefore, the 2D + 1 model assumes the following form:



3.26


( ) ( ) ( ) ( ) ( ) z x q z x q t x q
z
T
z k
z x
T
z k
x t
T
c
L M BL
, , , + + =
|
.
|

\
|
c
c
c
c
÷
|
.
|

\
|
c
c
c
c
÷
c
c
p
85
where q
L
is the heat from the laser source, q
M
is the volumetric metabolic heat, q
BL
is the heat
due to mean blood flow speed u
BL
in the plane of vessels, k(z) is the thermal conductivity of a
particular layer, while ρ and c are the tissue density and specific heat respectively. In the model,
the following boundary conditions are used:





3.27



λ is the convection heat transfer coefficient, which depends on air flow. According to [110]:
λ = 2.7 + 7.4 (v
air
)
0.67
(W/m
2
K), where v
air
is the air speed in (m/s), q
ir
is the radiation heat flux:
q
ir
= σ (T
epi
4
− T
air
4
), where σ is the Stefan-Boltzmann constant and is the skin emissivity.
The heat source term associated with blood flow is assumed to have the decomposition
q
BL
= u
BL
(t)r(x, z), where u
BL
is the unknown blood flow speed in the plane of vessels and r(x, z)
the modified bell function:



3.28

h
v
is the thickness of the plane of vessels seen as a heat source. µ is defined as follows:


3.29

where ρ
BL
and c
BL
are the density and the specific heat of blood respectively, A is the vessels
cross section, and V is the control volume of tissue. We assume that the early temperature of
the blood in the vessels is the same as the core (dermis) temperature T
vessels
=T
dermis
.
The heat source term associated with laser source is assumed to have a gaussian expression
( )
( ) ( ) L x t x T t x T
L x T t D x T
epidermis
core
, 0 , ) , 0 , (
, 0 ) , , (
e =
e =
( ) ( ) ( )
( ) D z
x
t z L T
L x q T t x T
z
t x T
ir air
, 0 0
) , , / 0 (
, 0 , 0 ,
) , 0 , (
e =
c
c
e + ÷ =
c
c
ì
( ) ( ) ( ) t z x T t z x T
V
A
c
vessels BL BL
, , , , ÷ = p u
( )
( )
|
|
.
|

\
|
÷
÷ =
2
2
exp ,
v
h
d z
z x r
t
u
86

( )
2
) ( ) (
0
z w x z z L
e e q q
÷ ÷
=
o

3.30


with

Where q
0
is the central maximum depending on the laser fluence, is the total attenuation
coefficient depending on the biological layer,
a
is the absorption coefficient,
s
is the
scattering coefficient, g is the anisotropy coefficient, w(z) is the waist size, w
0
is the half-spot
diameter, is the laser wavelenght and t
p
is the pulse period.
In constructing this model, it has been assumed that there isn’t heat flux between the domain of
interest (0, L) × (0,D) with the rest of the body. This problem is well posed and has a unique
continuous solution. The solution is C
1
but has second order derivative in the z direction with
finite jump at the line of discontinuities z = C
t
of the thermal conductivity k.


3.5.1 Model Computation

In order to obtain a direct simulation of the model it has been used a Finite Volume (FV)
approximation with centered cells of size h
x
× h
z
on a regular space grid.
The finite volume method is one of the methods used for representing and evaluating partial
differential equations in the form of algebraic equations. In this method the values are
calculated at discrete places on a meshed geometry. "Finite volume" refers to the small volume
surrounding each node point on a mesh. In the finite volume method, volume integrals in a
partial differential equation that contain a divergence term are converted to surface integrals,
using the divergence theorem. These terms are then evaluated as fluxes at the surfaces of each
finite volume. Because the flux entering a given volume is identical to that leaving the adjacent
volume, these methods are conservative. The main advantage of the finite volume method is
that it is easily formulated to allow for unstructured and dishomogeneus meshes (structures).
Let us denote T
i,j
the average value T of in the centered FV cells. The discrete version of
equation (3.26) is:

3.31
{
0 =
L
q
P
t t >
P
t t s < 0
( ) ( ) | ( )
( )
( )
|
M z w x z z
VASO
j
n
j i BL
BL BL
n
j i
n
j i
BL BL
n
j i
n
j i
BL BL
n
j i
n
j i
q e e q
D
h z
T u
c
z x t
c
x t
c
z t
z x T z x T + +
|
|
.
|

\
|
÷
÷ ÷
A A A
= u ÷ u
A A
÷ u ÷ u
A A
÷ A A ÷ A A
÷ ÷ + +
÷
+
+
+
÷
+
+
+
2
) ( ) (
0
2
2
1
,
1
2 / 1 ,
1
2 / 1 ,
1
, 2 / 1
1
, 2 / 1 ,
1
,
exp 1 ) (
o
t
u
p p p
( ) g
s
÷ + = 1
a
o o o
|
|
|
.
|

\
|
|
.
|

\
|
+ =
2
2
0
1
2
0
2
) ( w z w z w t ì
87
In the cell centred in (x
i
,z
j
),


3.32


The heat flux values at the wall of the cells are approximated with :




3.33


and



3.34

Where, in the equation (3.31), T
n
i,j
denotes the temperature at time ndt.
Using the boundary conditions on Φ and T we obtain a classic linear system:

3.35

At each time step we have to solve a linear system to obtain T
n+1
(x,z). M is pentadiagonal
matrix of size (N
x
× N
z
)
2
. T
hx,hz
is a matrix of size N
x
× N
z
reshaped from the 2D unknown
solution’s array (T
i,j
), i=1..N
x
, j=1..N
z
column-wise or row-wise. The grid computation is fixed
to N
x
=N
z
=64 and the time step is of the order of 1/20 which corresponds to the time step
between two frames in our thermal camera.
The numerical computations have been realized using the Matlab 6.5 code. The main thermal
and optical costants used in the simulations, for the biological tissues, will be showed in the
next chapter.
z z
z
z j
x x
x
x i
N i h
N
j
h z
N i h
N
i
h x
.. 1
1
1
2 /
.. 1
1
1
2 /
=
÷
÷
+ =
=
÷
÷
+ =
1 .. 1
1 .. 1
, 1 ,
2 / 1 , 2 / 1 ,
, , 1
, 2 / 1 , 2 / 1
÷ =
÷
=
÷ =
÷
=
+
+ +
+
+ +
z
z
j i j i
j i j i
x
x
j i j i
j i j i
N j
h
T T
k
N i
h
T T
k
o
o
( )
( )
j i j i j i
j i j i j i
k k k
k k k
, 1 , 2 / 1 ,
, , 1 , 2 / 1
2
1
2
1
+ =
+ =
+ +
+ +
z
h
x
h
z
h
x
h
MT
, ,
o =
88
Chapter 4

EXPERIMENTAL AND NUMERICAL RESULTS

4.1 Introduction

In this last chapter it will be shown results obtained through the monitoring of patients affected
by plane angioma in the facial region. After a description of the instruments used (laser system
and infrared camera) and of the operative protocol realized the main results will be presented,
confirming the great contribute of our methodological approach.
First of all, in order to emphasize the thermal behaviours associated to differents facial region,
some preliminaries evaluations relative to temperatures and temporal dynamics found will be
illustrated. Subsequently, results obtained on different patients appling the approach described
in the previous chapter will be showed and discussed. Particularly, thanks to the pulsed
thermography model, depth and thickness relative to different plane of vessels have been
estimated and in each different clinical situations numerical simulations have been run.
Experimental and numerical values will be analized and compared. Starting from this results,
some set of laser parameters relative to plane angioma characterized by specific depth and
thickness, will be proposed. The chapter is completed with a discussion relative to some limit
conditions for laser treatment application on plane angioma pathology.

a) b)












Fig 4.1 : Example of laser treatment on patient with plane angioma: a) infrared image, b) visible image.

89
4.2 Materials and method

In the experimental activity of this work, the laser treatments on 26 patients affected by plane
angioma have been monitored. The 26 patients were 17 male and 9 female with age in the
range 20-50. In order to study an homogeneous sample group we monitored all patients with a
plane angioma in the facial region. In the majority of cases, the angioma was widespread in
different facial region and principally on cheek, nose and temple.
The laser therapy is applied in photothermal regime using the selective photothermolysis
principle. The laser system used has been Pulsed Dye Laser – Deka Dermobeam 2000 with
characteristics showed in fig 4.2


PDL - Deka Dermobeam 2000


Display Laser tool Handle


LASER
SYSTEM












Fig 4.2 : Main characteristics of the Pulsed Dye Laser – Deka Dermobeam 2000



As described in chapter 1, being haemoglobin the most relevant chromophore in vascular
pathologies, the Dye laser naturally becomes the elective therapeutic tool by selective
absorption. In fact, as it is possible to see from figure 1.3, the emission wavelength of this laser
Wavelegth 595 nm
Energy density (Fluence) 3 - 35 J/cm
2
Pulse duration 0.5 – 40 ms
Spot diameter 2 - 5 - 7 mm
Emission 15 Hz
Laser mode Gaussian
Energy stability in emission 5 %
90
(595 nm) is very close to the absorption peak of the target (haemogloblin) as required from
selective photothermolysis principle.
The infrared acquisition has been performed by means of two differents digital thermal Camera:
a Land FTI6 - HgCdTe camera used in preliminaries evaluation and a more performanced
Avio TVS 500 – microbolometer camera used in the whole study. The main characteristics of
the cameras are showed in the fig. 4.3. Particularly, for the Avio camera the acquisition
spectral band is within 8-14μm, the time resolution is 0.02 s, and the temperature sensitivity is
0.06 K.




INFRARED
CAMERAS



Land FTI6 Avio TVS500

Sensor InGaAs –FPA 256X256 Microbolometer –FPA 320X240
Spectral range 3 - 5 μm 8 - 14 μm
Temperature sensitivity 0.1 °C 0.06 °C
Image resolution 8 bit 14 bit
Max Frame Rate 15 Hz 60 Hz

Fig 4.3 : Main characteristics of the infrared cameras employed

Black body correction has been executed to evaluate and to correct drift and shift effects due to
the infrared camera at the beginning of each measurement session. Emissivity of the skin has
been estimated as =0.95. In the next pictures the system and laser therapy ambulatory are
shown:










Fig 4.4 : Laser theraphy ambulatory: infrared camera and laser tool
91
The medical protocol commonly used for the treatment of patients with plane angioma
pathology by the physician is the following:

- He chooses the area of the angioma that he wants to treat (of dimension about 5x5 cm)
- He defines the laser parameters (usually, starting from referential laser parameters in tab. 3.1)
- He covers the area chosen with single laser pulses with an interval of 2-4 s

Particularly, in the covering of the area of the patient the physician laids a pulse upon the other
of about 20 % of spot diameter in order to reduce dischromic effects.
In order to apply the operative approach described in the previous chapter, the next procedure
has been performed sequentially:

- Acclimatization of the subject in the therapy rooms for a time not less than 15 minutes
- The anatomic facial region of intereset has been cleaned with alcohol in order to remove
sweat, cosmetics, dust and others
- A single pulse with referential laser parameters has been applied in the middle of the
region of interest
- An infrared recording of the region under investigation has been made continuously some
seconds before, during and after (for a time of about 30 s) the laser pulse.

The laser therapy room has been thermostated at 24 0.5 °C during the measurement. The
umidity was about 60%.
This procedure has permitted the registration of the thermal behaviour of the specific anatomic
region under laser-interaction: temperature and its temporal evolution (thermal recovery).
Subsequently, each information has been analyzed with the PT model and with the numerical
simulation described in chapter 3. In the next sections the results achieved will be showed and
discussed starting from some preliminaries evaluations. For the data and numerical analysis,
self-implemented software has been realized with a MATLAB platform.








92
0 5 10 15 20 25 30
35
40
45
50
55
60
65
70
P5
P4
P3
P2
P1
T
e
m
p
e
r
a
t
u
r
e

(
°
C
)
Event
Cheek
T
i
T
F
4.3 Preliminary results and observations

The first step was aimed to evaluate the thermal responses of different parts of the skin
illuminated by laser power. In this section, some graphics of thermal response relative to
representatives cases, extrapolated from the monitoring on all patients, are reported with the
intent of emphasizing the main results obtaind from these preliminaries evaluation. All
graphics reported in this section refer to a set of referential laser parameters shown in the next
table:














Tab. 4.1 : Set of laser parameters used in the preliminaries monitoring

In the first graph temperatures obtained in a specific anatomic region after a single laser pulse
are compared for different patients















Fig. 4.5 : Temperatures obtained after laser pulse refered to cheek region
Referential laser parameters

Wavelenght

595 nm

Energy density

7 J/cm2

Pulse duration

1.5 ms

Spot diameter

7 mm

T 25 °C

93
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
36
40
44
48
52
56
60
64
Event
Cheek
Nose
Temple
T
e
m
p
e
r
a
t
u
r
a

(
°
C
)
In the graph with black and red point we indicated the initial and final temperatures recorded
before and after the laser pulse respectively. The evaluations refer to the same facial region
affected by plane angioma (cheek) for 5 different patients (from P1 to P5). For each patient six
thermal response associated to laser parameters in table 4.1 are graphicated.
As it is possible to observe, even if the initial temperatures are quite similar for all patients
(physiological temperature), the final temperature, obtained immediately after the pulse, can be
very different.
















Fig. 4.6 : Temperatures obtained after laser pulse refered to cheek region



Similar situation has been found for the thermal response obtained in different facial region of
a same patient affected by plane angioma. In general, from the evaluations performed on all
patients it has been found with the same laser parameters, on a same patient and in presence of
the same pathology, that the T among differents anatomics regions can be higher then
15-20 °C. For example, fig. 4.6 which refers to a patient affected by widespread plane angioma,
shows how the temperature obtained immediately after the pulse can change from about 52 °C
on the cheek to 64 °C on the tample.
However, both fig. 4.5 and 4.6, for a same patient in a same facial region show a good
repetitions of the temperatures measured. The temporal evolutions of the temperature after the
laser pulse has been recorded and analyzed also for different anatomics regions. In general, this
dynamic depends on the region under investigation. Fig. 4.3 shows how each anatomic region
of the face (of a same patients) is characterized from a different recovery time (Tr).

T 20 °C

94
-0,5 0,0 0,5 1,0 1,5 2,0 2,5 3,0 3,5 4,0
32
36
40
44
48
52
56
60
Nose
Cheek
Temple
T
e
m
p
e
r
a
t
u
r
e

(
°
C
)
Time(s)
The recovery time is defined as the time occurring by the tissue for recovering the 70% of the
induced thermal gap.


















Fig. 4.7 : Temporal evolutions of the temperature after a laser pulse in three region of the face

This dependence has been found in each patient with widespread plane angioma where it has
been possible to realize evaluations in different regions affected by pathology. In next table the
average recovery times evaluated on three facial region are reported

















Tab. 4.2 : Mean values of the recovery time estimated in three different facial regions

Recovery time for differents facial regions

Region

Tr (s)

Temple

2.4 0.5

Nose

4.1 0.6

Cheek

6.2 0.6
95
0 2 4
36
40
44
48
52
56
60
64
68
T
e
m
p
e
r
a
t
u
r
e

(
°
C
)
Time (s)
The data in table 4.2 refer to mean values obtained on all recovery time estimated in the
temporal evolutions recorded.
As well as for the temperatures, when the analysis is focused on a same patients in a same
anatomic region a good repeatability of the dynamics has been found. In fig. 4.8 temporal rate
of the temperature refered to the cheek of a patient and estrapolated from a recorded laser
treatment are reported.


















Fig. 4.8: Temporal rate of the temperature refered to the cheek of a patient after application of laser pulses


The graph shows four dynamics associated to four single laser pulse made in sequence on four
neighbouring points of the tissue. The pulses are numerated from 1 to 4 in temporal order.
As it is possible to note all temporal rate reported are quite similar. In the graph, the initial
temperatures refered to the dynamics are different. This is due to the thermal superimposition
effect present when the time awaited between two successive pulse is lower than the time spent
from the tissue for a complete recovery.

All evaluations shown since here, emphasise how the thermal behaviours of the tissue achieved
under laser treatment for a given set of laser parameteres, depend on the specific target.
Particularly, they can differ from region to region and from patient to patient.
It is important remember, as it has been discussed in chapter 1 that in phototermal interaction
regime, the temperature induced to the tissue is strictly linked to hystological modification
1
2
3
4
96
obtained. Therefore, these preliminary results point out how with the same laser parameters,
changing patient or anatomic region, the effects and so the therapeutic efficiency obtained can
greatly differ, changing from optimal therapeutic results to unwanted harmful effects (for
istance scars formation). These observations justifie and confirm thtat if not using specific and
tailored laser parameters for each different clinical situation the laser treatments on plane
angioma pathology permits an optimal improvement just in a small number of patients, as
discussed in the section 3.3. Starting from this observations, in order to improve the therapeutic
efficiency and the safety in the laser therapy, the request of optimized set of laser parameters,
for each specific treatment, becames a necessity. With this intent the operative approach, object
of this thesis work and described in previous chapthers, has been performed and the main
results achieved will be shown and discussed in the next section.

































97
4.4 Pulsed Thermography model: experimental results

The operative procedure described in the section 4.2 has been applied to all patients in the
anatomics regions of the face where the angioma was present. In this regions, the temporal
evolution of the temperature achieved after a laser pulse has been recorded for each clinical
situation. Successively, the recorded dynamics have been analized with the pulsed
thermography model descibed in the section 3.4. For each patient through this model the depth
and thickness of the plane of vessels associated to plane angioma in the region under treatment
has been estimated . The results obtained for each patients are shown in table 4.3

Patient Region Depth (μm) Thickness (μm) Patient Region Depth (μm) Thickness (μm)

1
cheek 210 10 80 20
14
cheek 660 100 240 130
temple 120 10 80 20 temple 390 40 160 60
nose 150 10 70 20 nose 450 40 220 120
2 cheek 310 30 110 40 15 cheek 630 100 340 240
temple 160 10 70 20 temple 510 80 220 120
3 cheek 170 10 70 20 16 cheek 600 90 310 210
nose 140 10 80 20 nose 540 80 180 70
4 cheek 550 80 230 130 17 cheek 730 110 340 240
temple 520 80 150 60 temple 410 40 180 70

5
cheek 410 40 160 60 nose 480 40 250 130
temple 320 30 130 50 18 cheek 610 90 110 40
nose 330 30 150 60 temple 530 80 260 140
6 cheek 570 80 210 120
19
cheek 590 90 180 70
7 cheek 800 160 320 220 temple 320 30 160 60
nose 670 100 340 240 nose 520 80 230 120
8 cheek 595 90 250 130 20 cheek 900 180 370 250
temple 210 10 90 20 temple 770 150 310 210

9
cheek 350 30 310 210 21 cheek 230 10 80 20
temple 310 30 90 20
22
cheek 520 80 240 130
nose 330 30 170 70 temple 330 30 130 50

10
cheek 265 20 160 60 nose 510 80 210 110
temple 160 10 90 20 23 cheek 420 40 160 60
nose 190 10 110 40
24
cheek 810 160 280 150

11
cheek 430 40 220 120 temple 270 20 140 50
temple 260 20 80 20 nose 720 140 330 230
nose 400 40 180 70
25
cheek 780 150 280 150
12 cheek 470 40 140 50 temple 240 20 90 20

13
cheek 780 160 270 140 nose 710 140 310 210
temple 220 10 90 20 26 cheek 490 40 120 50
nose 470 40 250 130 nose 510 80 120 50

Tab. 4.3: Depth and thickness of the plane of vessels evaluated with the PT model
98
The evaluations in table 4.3 refer to mean values achieved on 5 different measurement. As it is
possible to see from the table, the majority of patients were affected from a plane angioma
widespread in more region of the face. In the next histograms the values for the depth and
thickenss of the angioma are reported without distinction of the anatomic region.


























Fig. 4.9: Distributions relative to the depth and thickness evaluated globally on all anatomic facial region




As it is possible to note, the depth estimated mainly changes in the range from 300 μm to 700
μm while the thickess changes from 100 μm to 300 μm. Despite the previous case, in the next
histograms the results refer to single anatomic facial region.





100 200 300 400 500 600 700 800 900
0
10
20
30
40
50
F
r
e
q
u
e
n
c
y

(
%
)

Depth ( m)
0 50 100 150 200 250 300 350 400
0
10
20
30
40
50
F
r
e
q
u
e
n
c
y

(
%
)
Thickness ( m)
99
0 50 100 150 200 250 300 350 400
0
5
10
15
20
25
30
35
40
45
50
Angioma area : Nose
F
r
e
q
u
e
n
c
y

(
%
)
Thickness ( m)



























Fig. 4.10: Distributions relative to the depth and thickness evaluated on three anatomic facial region


As shown in the histograms, in average, the deeper and thicker plane angioma has been found
in the cheek while the less depth and less thick on the temple. From these results, the formation
of plane angioma seems to spread preferentially the anatomic region of the cheek. This is
probably due to the higher presence in the cheek of subsurface fat respect to other facial region.
A higher presence of fat, probably, facilitates the extention of the angioma differently from the
other regions (for istance the temple) characterized by an higher presence of subsurface
muscular tissue. These data permit, for each clinical situation, a specific morphological
reconstruction of the multilayer biological tissue under treatment. Subsequently, on these
multilayer it has been executed the 2D+1 numerical simulation based on the biometric heat
transfer model described in chapter 3. In the simulations the thermal behaviour of the
biological multilayers has been achieved for different set of laser parameters as input. The
exerimental results and the simulations refered to differents biological multilayer are reported
and discussed in the next section.
100 200 300 400 500 600 700 800 900
0
10
20
30
40
50

Angioma area : Cheek
F
r
e
q
u
e
n
c
y

(
%
)

Depth ( m)
100 200 300 400 500 600 700 800 900
0
10
20
30
40
F
r
e
q
u
e
n
z
a

p
e
r
c
e
n
t
u
a
l
e

(
%
)

Profondità ( m)
Area angioma : Naso
100 200 300 400 500 600 700 800 900
0
10
20
30
40
Area angioma : Fronte
F
r
e
q
u
e
n
z
a

p
e
r
c
e
n
t
u
a
l
e

(
%
)

Profondità ( m)
0 50 100 150 200 250 300 350 400
0
5
10
15
20
25
30
35
40
Area angioma : Guancia
F
r
e
q
u
e
n
z
a

p
e
r
c
e
n
t
u
a
l
e

(
%
)
Spessore ( m)
0 50 100 150 200 250 300 350 400
0
10
20
30
40
F
r
e
q
u
e
n
z
a

p
e
r
c
e
n
t
u
a
l
e

(
%
)
Spessore ( m)
Area angioma : Fronte
100
4.5 Numerical simulations: results


The evaluations obtained in the previous section permit the reconstruction of the biological
multilayer made of: epidermis, dermis, plane of vessels and dermis again, associated to
different anatomic region of the face for each patient. Subsequently, on these multilayers it has
been performed a numerical simulation based on a biometric heat transfer model described in
the section 3.5. In the simulations the thermal behaviour of the biological multilayers has been
achieved for different set of laser parameters as input. The numerical output consist in the
evaluations of the spatials and temporal temperature profiles obtained on the multilayer after a
single laser pulse. Next, some results achieved from the simulations will be shown and
discussed. Moreover, different clinical situations will be compared among them. Starting from
these results, at the end of the section, some set of laser parameters will be proposed for the
treament of multilayers with specific plane angioma sizes. The numerical program of
calculation has been realized with a home-made Matlab code. In table 4.4 the thermal and
optical costants used in the simulations and adviced from literature [110-114] are shown .

Optical and Thermal costants
Epidermis Dermis Blood vessels

Density

1 Kg dm
-3

1 Kg dm
-3

1,055 Kg dm
-3

Specific heat

3,7 *10
-3
J Kg
-1
K
-1

3,7 *10
-3
J Kg
-1
K
-1


3,6 *10
-3
J Kg
-1
K
-1


Conductivity

0,209 *10
-6
W m
-1
K
-1

0,322 *10
-6
W m
-1
K
-1


0,492 *10
-6
W m
-1
K
-1


Absorption coefficient

18-60 cm
-1

2,2 cm
-1

147 cm
-1

Scattering coefficient

(g)

470 cm
-1

(0.790)

129 cm
-1

(0.790)

468 cm
-1

(0.995)

Refraction index
( =595nm)

1,37

1,37

1,33

Tab. 4.4: Optical and thermal costants used in the simulations

The optical costants in table 4.4 refer to wavelenght of 595 nm and patients of phototype III. In
the calculations the biological multilayer has been divided in a grid of 64x64 cells, the time
step between two frames has been fixed to 1/20 s and the depth of the epidermal layer has been
fixed to 100 m.
101
- Numerical solution: thermal dynamics associated to the problem

In the next images, a typical case of temporal evolution achieved is shown. The sequences refer
to the dynamics simulated on a structure with a plane angioma depth 600 μm and thick 100 μm
obtained after a single laser pulse. The laser parameters used as input are reported in the tab. 4.5










Fig 4.11: multilayer used in the simulation Tab. 4.5: Laser parameters used in the simulation

0 s 1 s 2 s 3 s







4 s 5 s 6 s 7 s






8 s 9 s 10 s 11 s







12 s 13 s 14 s 15 s





Fig 4.12: Temporal sequence of frames extrapolated from the simulations
laser parameters
Energy density - Fluence 7 J/cm
2
Pulse duration 5 ms
Spot diameter 7 mm
102
The images reported have been extrapolated from a full sequence achieved from the simulation.
The images corresponding to the istant t=0 s refers to the first grid calculated after the laser
pulse. As it is possible to note from the sequence the thermal dynamics in the layers isn’t
neither homogeneous nor symmetrical as it was expected. For example, the layers more
superficial recover quickly than deep layers and the heat, generated from the laser pulse, moves
towards the core of the structure. This is mainly due to the different boundary condition present
in the model (eq. 3.27) and associated to epidermis (z=0) and dermis (z=D, core) layers.
Moreover, as widely discussed, the biological structure is a dishomogeneous multilayer with
optical and thermal properties depending from the different tissues. In particular it is possibile
to note from fig. 4.12 (at t=0) how the temperature achieved on plane agioma layers is higher
than the temperature found for epiderimis or dermis because of the higher absorption
coefficient at laser wavelenght (595 nm) of the plane of blood vessels (table 4.4). This
particular dynamics has been found in all numerical solutions calculated in different situations
(for different multilayers and different laser parameters) and it is peculiar for this problem.
Changing the biological multilayers and the laser parameters, this dynamics is obviously
characterized from different temperatures for the layers and different recovery times.

















103
-8 -6 -4 -2 0 2 4 6 8
0
10
20
30
40
Length (mm)
C
Fluence = 7 J/cm
2
,
Spot diameter = 7 mm
t
pulse
= 5 ms
Plane Angioma
Epidermis
Epidermis - Experimental points
0 2 4 6 8 10 12 14
0
5
10
15
20
25
30
35
40
45
time (s)
C
Fluence = 7 J/cm
2
,
Spot diameter = 7 mm
t
pulse
= 5 ms
Plane Angioma
Epidermis
Epidermis - experimental points
- Numerical solutions refered to a same structure and for different laser parameters

In order to optimize the laser parameters for each specific treatment, the temperature obtained
on the plane of vessels and on the other tissues of the structure must be evaluated for different
set of laser parameters. In the next figures the results obtained from simulation on the
biological multilayer in fig. 4.11 and with the laser parameters of the tab. 4.5 are reported. In
particular, the distribution of the T induced on the structure calcolated at istant t=0
(immediately after the laser pulse), is showed in 2-D and 3-D rappresentation. Moreover, the
temporal evolutions and the spatial distribution of the temperature on the epidermis and on the
plane of vessels are compared.










Fig 4.13: Distribution of the temperature calcolated at istant t=0 s, 2D and 3D rappresetation












Fig 4.14: Temporal evolutions and spatial distribution of T on the epidermis and on the plane of vessels
104
0 2 4 6 8 10 12 14
0
10
20
30
40
50
60
C
time (s)
Fluence = 7 J/cm
2
,
Spot diameter = 7 mm
t
pulse
= 1 ms
Plane Angioma
Epidermis
Epidermis - Experimental points
-8 -6 -4 -2 0 2 4 6 8
0
10
20
30
40
50
60
Length (mm)
C
Fluence = 7 J/cm
2
,
Spot diameter = 7 mm
t
pulse
= 1 ms
Plane Angioma
Epidermis
Epidermis - Experimental points
In the graphs in fig. 4.14 the temperatures for the plane angioma and the epidermis estimated
from the simulation have been rappresented respectively with black and red point. In the same
graphs, the experimental values recorded on the epidermis with the infrared camera are
reported with green points. It is possible to observe as this last are in a good agreement with
simulated data. The induced thermal gap calculated between the plane angioma and epidermis
at istant t=0 s has been ~ 15 °C.
Starting from these laser parameters, the next graphs show how the numerical solutions change
when the spot diameter moves from 7 mm to 2 mm or the pulse duration from 5 ms to 1 ms.














Fig 4.15: Distribution of the temperature calcolated at istant t=0 s, 2D and 3D rappresetations










Fig 4.16: Temporal evolutions and spatial distribution of T on the epidermis and on the plane of vessels
laser parameters
Fluence 7 J/cm
2
Pulse duration 1 ms
Spot diameter 7 mm
105














Fig 4.17: Distribution of the temperature calcolated at istant t=0 s, 2D and 3D rappresetations











Fig 4.18: Temporal evolutions and spatial distribution of T on the epidermis and on the plane of vessels

These last graphs show how, at istant t=0 s, both smaller spot diameter and shorter pulse
duration induces a thermal gap between epidermis and plane angioma higher than the previous
case. In fact, the difference on the maximum temperature achieved changes from ~ 15 °C to ~
25 °C when the spot diameter changes from 7 mm to 2 mm and from ~ 15 °C to ~ 45 °C when
the pulse duration changes from 5 ms to 1 ms. In these cases, a good agreement between
numerical solutions and experimental data refered to epidermis layers has been obtained.
laser parameters
Fluence 7 J/cm
2
Pulse duration 5 ms
Spot diameter 2 mm
0 2 4 6 8 10 12 14
0
10
20
30
40
50
60
70
80
90
100
time (s)
C
Fluence = 7 J/cm
2
,
Spot diameter = 2 mm
t
pulse
= 5 ms
Plane Angioma
Epidermis
Epidermis - experimental points
-8 -6 -4 -2 0 2 4 6 8
0
20
40
60
80
100
C
Length (mm)
Fluence = 7 J/cm
2
,
Spot diameter = 2 mm
t
pulse
= 5 ms
Plane Angioma
Epidermis
Epidermis - Experimental points
106
0 2 4 6 8 10 12 14
0
10
20
30
40
50
60
Plane Angioma
Spot diameter = 7 mm , t
pulse
= 5 ms
Fluence = 4 J/cm
2
Fluence = 7 J/cm
2
Fluence = 10 J/cm
2
C
time (s)
In the next numerical results the dynamics of the thermal gap induced by the laser pulse on the
biological structure of fig. 4.11 are compared when two of the three laser parameters are fixed
and just one is changed. A 2-D and a 3-D rappresentation is given in all three different
situations. In the first case, the pulse duration and spot diameter have been fixed while three
different values for the fluence have been used.





4 J/cm
2
7 J/cm
2
10 J/cm
2













Fig 4.19: Distribution of the temperature calcolated at istant t=0 s, 2D and 3D rappresetations









Fig 4.20: Temporal evolutions of T on the plane of vessels fordifferent fluence
laser parameters
Pulse duration 5 ms
Spot diameter 2 mm
107
0 2 4 6 8 10 12 14
0
5
10
15
20
25
30
35
40
45
C
time (s)
Plane Angioma
Fluence = 4 J/cm
2
, Spot diameter = 7
t
pulse
= 1 ms
t
pulse
= 5 ms
The graph in fig. 4.20 shows clearly how the induced thermal gap can change from ~ 25 °C to
~ 60 °C for the range of fluence used in these simulations (4-10 J). In the second case, the
fluence and spot diameter have been fixed while two different values of the pulse duration have
been used.



















Fig 4.21: Distribution of the temperature calcolated at istant t=0 s, 2D and 3D rappresetations









Fig 4.22: Temporal evolutions of T on the plane of vessels fordifferent pulse duration
laser parameters
Fluence 4 J/cm
2
Spot diameter 2 mm
108
0 2 4 6 8 10 12 14
0
10
20
30
40
50
60
70
80
time (s)
C
Plane Angioma
Fluence = 4 J/cm
2
, t
pulse
= 5 ms
Spot diameter = 2 mm
Spot diameter = 7 mm
In the last case, in the calculations, two values for the spot diameters have been used while the
others two parameteres have been fixed





















Fig 4.23: Distribution of the temperature calcolated at istant t=0 s, 2D and 3D rappresetations









Fig 4.24: Temporal evolutions of T on the plane of vessels fordifferent pulse duration
laser parameters

Fluence

7 J/cm
2

Pulse duration

5 ms
109
As it is possibile to observe from the last graph, the induced thermal gap from the laser pulse
can vary massively when the spot diameter is changed.


- Numerical solutions refered to different structures with the same laser parameters

In this paragraph the results relative to the biological multilayer with a plane angioma at
different depths or with different thicknesses are shown. All simulations and results shown in
this section refer to the laser parameters reported in the table 4.5.
The first case is relative to three structures with a plane of vessels at different depth but with
the same thickess. These structures associated to three different clinical situations have been
numerically analysed. The depth of the angioma inside the three structures was respectively
300 μm , 600 μm and 900 μm while the thickness was 100 μm.












Fig 4.25: multilayer used in the simulations














Fig 4.26: Distribution of the temperature calcolated at istant t=0 s, 2D rappresentation.

110






























Fig 4.27: Temporal evolutions and spatial distribution of T on the epidermis and on the plane of vessels


As in the previous case the false colours images extrapolated from the numerical solutions refer
to the istant subsequent to the laser pulse (t=0 s). From the images in fig. 4 the temporal
evolution and the spatial distribution of T, in case of both epidermis and plane angioma layers,
are graphicated. The simulations show how the difference found on the temperatures on
structures affected by plane angioma to different depth in the tissue, are higher on the plane
angioma layers than epidermis layers. For epidermis, T change in the range from ~23 °C to ~
40 °C, with a gap of ~17°C, and for plane angioma layers from ~25 °C to ~75 °C, with a gap of
~50°C. As it is expected, the temperatures obtained on the plane angioma are higher than the
temperatures on the epidermis but the ratio T
angioma
/ T
epidermis
decreases for deeper plane of
vessels. As it will be discussed later, this trend restricts just to patients affected by plane
0 2 4 6 8 10 12 14
0
5
10
15
20
25
30
35
40
Fluence = 7 J/cm
2
,
Spot diameter = 7 mm
t
pulse
= 5 ms
Thikness = 100 m
Depth = 900 m
Depth = 600 m
Depth = 300 m
time (s)
C
Epidermis
-8 -6 -4 -2 0 2 4 6 8
0
5
10
15
20
25
30
35
Fluence = 7 J/cm
2
,
Spot diameter = 7 mm
t
pulse
= 5 ms
Thikness = 100 m
Depth = 900 m
Depth = 600 m
Depth = 300 m
Length (mm)
C
Epidermis
0 2 4 6 8 10 12 14
0
20
40
60
80
time (s)
C
Plane Angioma
Fluence = 7 J/cm
2
,
Spot diameter = 7 mm
t
pulse
= 5 ms
Thikness = 100 m
Depth = 900 m
Depth = 600 m
Depth = 300 m
-8 -6 -4 -2 0 2 4 6 8
0
20
40
60
80
100
C
Plane angioma
Fluence = 7 J/cm
2
,
Spot diameter = 7 mm
t
pulse
= 5 ms
Thikness = 100 m
Depth = 900 m
Depth = 600 m
Depth = 300 m
Length (mm)
111
angioma which are not too deep. In this case a laser therapy can supply good results and
reduced risks.
In the next case analized, the numerical solutions relative to two biological structures with a
plane angioma at same depth but with different thicknesses have been compared. For the two
structures the thickness was 100 μm and 300 μm respectively, while the depth was 600 μm.





































Fig 4.28: Distribution of the temperature calcolated at istant t=0 s, 2D rappresentation


112
-8 -6 -4 -2 0 2 4 6 8
0
5
10
15
20
25
30
35
40
Length (mm)
C
Epidermis
Fluence = 7 J/cm
2
,
Spot diameter = 7 mm
t
pulse
= 5 ms
Depth = 600
Thikness = 300 m
Thikness = 100 m






























Fig 4.29: Temporal evolutions and spatial distribution of T on the epidermis and on the plane of vessels


Differently from the previous case, from the graphs reported in fig. 4.29 the variations on the
temperatures achieved in the two situations for the plane angioma and for epidermis are,
substantially, the same. In fact, the maximum T found for the plane angioma goes from
~42 °C to ~52 °C while the maximum T found for the epidermis goes from ~ 28 °C to
~37 °C, with a gap in both situations of about ~9-10°C. These results demonstrate how the
thermal behaviour achieved in the multilayer biologial structures affected by plane angioma is
more sensitive to the variations of the depth than variations of the thickness.
In the next graphs the ricovery time and the induced thermal gap ( T) versus the depth of the
plane angioma in the tissue are reported. The data refer to evaluations achieved in the
simulations. The values are relative to plane angioma with thickness 100 μm and 300 μm and
various depths.
0 2 4 6 8 10 12 14
0
5
10
15
20
25
30
35
C
Epidermis
Fluence = 7 J/cm
2
,
Spot diameter = 7 mm
t
pulse
= 5 ms
Depth = 600 m
Thikness = 100 m
Thikness = 300 m
time (s)
0 2 4 6 8 10 12 14
0
10
20
30
40
50
time (s)
C
Plane Angioma
Fluence = 7 J/cm
2
,
Spot diameter = 7 mm
t
pulse
= 5 ms
Depth = 600 m
Thikness = 100 m
Thikness = 300 m
-8 -6 -4 -2 0 2 4 6 8
0
10
20
30
40
50
60
C
Plane Angioma
Fluence = 7 J/cm
2
,
Spot diameter = 7 mm
t
pulse
= 5 ms
Depth = 600
Thikness = 300 m
Thikness = 100 m
Length (mm)
113
300 400 500 600 700 800 900
20
30
40
50
60
70
80
90
100
C
Epidermis
Plane Angioma
Fluence = 7 J/cm
2
,
Spot diameter = 7 mm
t
pulse
= 5 ms
Thickness = 100 m
Thickness = 300 m
Experimental points
Depth ( m)
300 400 500 600 700 800 900
1
2
3
4
5
6
7
Depth ( m)
R
i
c
o
v
e
r
y

t
i
m
e

(
s
)
Fluence = 7 J/cm
2
,
Spot diameter = 7 mm
t
pulse
= 5 ms
Plane Angioma
Epidermis
Thickness = 100 m
Thickness = 300 m
Thickness = 100 m - Experimental value

















Fig 4.30: Recovery time versus depth for two different thickness




















Fig 4.31: Induced thermal gap versus depth for two different thickness


From the graphs, it’s clear, how the ricovery time increases with the depth while it doesn’t
change at all to variations of the thickness. Differently from the previous case, the T
decreases with depth and increases with the thickness of the plane of vessels. In the graphs the
experimental values refered to epidermis and extrapolated from some patients monitored are
reported with green points. The graphs show clearly a good agreement between simulated data
and experimental values in both situations.
114
0 2 4 6 8 10 12 14
0
5
10
15
20
25
30
35
40
45
50
55
C
time (s)
Fluence = 7 J/cm
2
,
Spot diameter = 7 mm
t
pulse
= 5 ms
Phototype I
Plane angioma
Epidermis
- Numerical simulation for different phototypes

All numerical simulations and results up to now reported refer to patients of phototype III
(typology more frequent). In this study, the thermal effects achieved on different phototypes
have been compared. From an optical point of view, patients of different phototype are
characterized from a different absorption coefficient for the epidermal layers. In fig. 4.33,
numerical simulations on three equal multilayers (with same dimensions for the plane of
vessels) but with different phototypes are shown. The results refer to structures with a plane
angioma depth 600 μm and thick 100 μm (fig. 4.11) and laser parameters reported in table 4.5.
The absorption coefficients for the epidermis used in the simulations are reported in table 4.6







Tab. 4.6: Absorption coefficients for the epidermis used in the simulations


Ph I Ph III Ph VI










Fig 4.32: Distribution of the temperature calcolated at istant t=0 s for different phototype












Fig 4.33: Temporal evolutions of T on the epidermis and on the plane of vessels for different phototypes
Absorption coefficient
Phototype I 7 J/cm
2
Phototype III 5 ms
Phototype IV 7 mm
0 2 4 6 8 10 12 14
0
5
10
15
20
25
30
35
40
45
50
55

C
time (s)
Fluence = 7 J/cm
2
,
Spot diameter = 7 mm
t
pulse
= 5 ms
Phototype III
Plane Angioma
Epidermis
0 2 4 6 8 10 12 14
0
5
10
15
20
25
30
35
40
45
50
55
C
time (s)
Phototype VI
Fluence = 7 J/cm
2
,
Spot diameter = 7 mm
t
pulse
= 5 ms
Phototype VI
Plane Angioma
Epidermis
115
-2 0 2 4 6 8 10 12 14
0
5
10
15
20
25
30
35
40
45
50
55
Pht III
Pht I
Pht VI
Fluence = 7 J/cm
2
,
Spot diameter = 7 mm
t
pulse
= 5 ms
Plane angioma
Phototype VI
Phototype III
Phototype I
time (s)
C
-2 0 2 4 6 8 10 12 14
0
5
10
15
20
25
30
35
40
45
50
55
Pht I
Pht III
Pht VI
C
time (s)
Fluence = 7 J/cm
2
,
Spot diameter = 7 mm
t
pulse
= 5 ms
Epidermis
Phototype VI
Phototype III
Phototype I
In figure 4.33 the temporal rate of the T calculated on epidermis and on plane angioma has
been compared for each different phototype. From the graphs, the thermal gap between
epidermal layer and plane angioma layer increases going from phototype VI to phototype I and
in particular, for phototype VI the laser pulse induces a maximum T on the epidermis
(for t= 0 s) higher than the maximum T obtained on the angioma. In next figure the temporal
rate of T refered to the epidermis and plane angioma are reported separately for the three
phototypes















Fig 4.34: Temporal evolutions of T on the epidermis and on the plane of vessels for different phototypes



The graphs show, clearly, as going from phototype I to phototype VI that the temperature
obtained on the epidermis increases while the temperatures found on the angioma decreases.
For the epidermis, the maximimum gap of the temperature has been estimated to be about
~37 °C (from ~18°C for phototype I to ~55°C for phototype VI) while for the angioma it has
been found of ~13 °C (from ~37 °C for phototype VI to ~50 °C for phototype I).











116
4.6 Limit conditions for applications of laser treatments


As discussed in chapter 1, in the laser therapy performed in the photothermal regime, the
temperatures induced on the tissues are strictly linked to the histological modification obtained.
In general, the goal of the laser therapy is to obtain a local and selective interaction with the
target structures (selective photothermolysis principle). Particularly, for treatment on plane
angioma, a complete photocoagulation of the blood vessels and, in the same time, a lower
interaction with the surronding tissues (dermal and epidermal layers) is desired. It means,
pratically, to maximize the temperature on the plane angioma and minimize the temperatures
on the epidermis and dermis induced by a laser pulse.
The last results in fig. 4.33 and 4.34 emphasize how in some circumstance this is impossible.
In fact, if the ratio S
I
= T
epidermis
/ T
angioma
can be regarded as a selectivity index of the
interaction, then for the tissues of phototype V-VI (where S
I
1) the application of laser
therapy can be hazardous.
In this patient the lower temperatures achieved on the angioma makes the blood vessels
photocoagulation improbable and at the same time the higher temperatures achieved on the
epidermis and dermis can produce the formation of scars due to the collagene denaturation. It is
possible to note, as the same situation (S
I
1) is found for angiomas deeper than about ~900
m (as it has seen above) independently from the phototypes, and, so, in these case too, the
results achieved by simulations advises against the application of the laser treatments.
















Fig 4.35: Distribution of the temperature calcolated at istant t=0 s for a phototype III
with a plane agioma depth 900 m and thick 100 m.

T
DERMIS
/ T
ANGIOMA
> 1

117
4.7 Final conclusions: set of laser parameters



The results shown up to now together with the complete study on the patients, give a wide
vision about the the thermal behaviour of the biological multilayer affected by plane angioma
when they are subject to selective laser interactions. These results represent a first step in order
to optimize laser parameters for specific clinical situations. In fact, as an output of this work it
comes that it is possible to choose a set of laser parameters in order to accomplish these
relations for each different biological multilayer











The above ranges, are chosen in oder to photocoagulate the blood vessels of the angioma and at
the same time to reduce the risks of scars formations. According to these indications, different
set of laser parameters can be proposed. Just as an example, in tab. 4.7 some set of laser
parameters are proposed for treatment of patients affected by plane angioma of different
dimensions. The set shown refers to patients of phototype III.
These parameters are given with the aim to achieve a more specific and appropriate interaction
with the tissue of the patient and to make laser treatments more safety and more efficient. With
this approach, others set can be suggested for different phototype and for different sizes of the
plane angioma.







T
EPIDERMIS,
T
DERMIS
< 30°C

45 °C < T
ANGIOMA
< 75 °C
Safety
Photocoagulation range
118

Tab 4.7: Some set of laser parameters proposed for a phototype III and for differents
dimesions of the plane angioma

Thickness
Depth

100 μm


300 μm





300 μm


Fluence = 3 J/cm
2

Spot diameter = 7 mm
Pulse duration = 5 ms

Fluence = 3 J/cm
2

Spot diameter = 7 mm
Pulse duration = 7 ms




600 μm


Fluence = 9 J/cm
2

Spot diameter = 7 mm
Pulse duration = 5 ms

Fluence = 5 J/cm
2

Spot diameter = 7 mm
Pulse duration = 5 ms





800 μm


Fluence = 16 J/cm
2

Spot diameter = 7 mm
Pulse duration = 5 ms

Fluence = 11 J/cm
2

Spot diameter = 7 mm
Pulse duration = 5 ms
119
In conclusion, with the approach described a good agreement between numerical solutions and
experimental results has been obtained for both ricovery time and temporal evolution of T in
the majority of cases, particularly, for more superficial angioma. In case of depeer angioma
(d > 900 m), a mismatch between numerical solution and experimental values have been often
found due to a higher error achieved in the evaluation of the dimensional parameters of the
angioma with the PT model used.
However, the results up to now obtained in this study are encouraging and a further goal is to
extend the analysis from a relatively small number of patients monitored to a larger population.
Therefore, in the next steps of this research activity, new PT models will be tested in order to
achieve a more accurate evaluation of the dimensional parameters of the plane angioma and,
subsequently, the set proposed will be gradually tested on different targets with the aim to
obtain, eventually, a complete validation of this novel approach.






















120
General Conclusions





In this thesis the laser – biological tissue interaction has been studied and analized with the
support of the infrared imaging technique.
This work started with an overview on the fundamental theoretical concepts of the laser–tissue
interaction. In particular, the main laser-tissue interaction mechanisms as the photothermal
interaction, the photochemical interaction, the photoablation, the plasma-induced ablation and
the photodisruption have been described. Greatest attention has been given to the description of
the photothermal regime of interaction, to the thermodynamic effect, and to the selective
photothermolysis principle which are core of the present work.
The state of art of the laser application in the medical field has been made in order to have an
overview of the different use of lasers in medicine, of the laser systems applied, of the
methodological approach commonly used and protocols. At the same time the results and
limits of the therapy have been estimated.

Subsequently, the principles and applications of the infrared imaging technique have been
reported. Both passive and active approach have been explained, mainly focusing pricipally the
attention on the different active procedures.
Several active thermographic techniques commonly used, that differ one another mainly in the
way data is acquired and/or processed, have been reviewed and discussed: pulsed approach,
step heating or long pulse method, lock-in thermography and vibrothermography. In this
description more emphasis has been given to the presentation of the pulsed termography
method because it is a part of the experimetal procedure used in this activity. Different
definitions of thermal contrasts used for inversion algorithms have been given and compared:
contrast for the transmission mode, contrast for the reflection mode and the differential
absolute contrast (DAC). The problems relative to the atmospheric trasmittance and infrared
sensors have been treated. An overview on the infrared detector as bolometers, pyroelectric
detectors, quantum detectors, Focal Plane Arrays (FPA) devices and on the their main
performance parameters has been made. Moreover, the novel approach and the new
applications of infrared imaging in medical science has been discussed.
121
In the third chapter, the core of the research activity has been described. The novel approach
proposed for the laser parameters optimization in the plane angioma treatment has been
illustrated and explained in details. The pathology analized (plane angioma) and computation
tools used have been treated and discussed. Particularly, the pulsed thermography model and
the numerical approach adopted in the operative protocol performed have been explained
rigorously.

The experimental activity was constituted of the following steps:

1) Assembly of the monitoring station including a thermal camera, a data acquisition
system and a laptop to elaborate the IR images collected.

2) Introduction of the monitoring station in the operative unit of photodynamic and
phototherapy ambulatory of the University of Naples "II Policlinico". The research
activity started, firstly, with a series of preliminaries clinical tests on patients affected
by Plane Angioma.

3) From the results of the preliminaries test a standard procedure for data acquisition has
been conceived (as described in section 4.2). All patients have been monitored along a
cycle of therapies of about 18 months. During this time different data refering to the
thermal behaviour of the tissue under treatment have been recorded.

4) The data recorded have been processed with the pulsed thermography model and then
with the numerical simulation based on a biometric heat transfer model according to
the operative approach described in the chapter 3.


The main experimental and numerical results found with this approach have been reported and
discussed in detail. First of all, the results relative to preliminaries test and observations have
been illustrated. In particular, it has been shown how comparing the thermal response obtained
in a specific anatomical region after a single laser pulse on different patients (with the same
laser parameters), the final temperature, achieved immediately after the pulse, can be very
different, with a T 25 °C too. A Similar situation has been found for the thermal response
obtained in different facial region of a same patient affected by plane angioma. In general,
122
from the evaluations performed on all patients it has been found how with the same laser
parameters, on a same patient and in presence of the same pathology, the T among different
anatomical regions can be higher than 15-20 °C. However, for a same patient in a same facial
region it has been obtained a good repetitions of the temperatures measured. The temporal
evolution of the temperatures after the laser pulse has been also recorded and analyzed for
different anatomical regions. In general, this dynamic depends on the anatomical region under
investigation and different average recovery times have been evaluated for each facial region.
However, as well as for the temperatures, when the analysis is focused on a same patient in a
same anatomical region a good repeatability of the dynamic has been achieved.

All evaluations obtained in the preliminary tests, emphasise how the thermal behaviours of the
tissues achieved under laser treatment for a given set of laser parameteres, depend on the
specific target. Particularly, they can differ from region to region and from patient to patient.
As reported, in the phototermal interaction regime, the temperature induced in the tissue is
strictly linked to the hystological modification obtained. Therefore, these results point out that
with the same laser parameters, changing patient or anatomical region, the effects and so the
therapeutic efficiency obtained can differ a lot, changing from optimal therapeutic results to
unwanted harmful effects (for istance scars formation). These observations justify and confirm
if we don’t use specifics and tailored laser parameters for each different clinical situation the
laser treatments on plane angioma pathology permits an optimal improvement just in a small
part of patients.

Subsequently, the results obtained using the novel approach proposed have been illustrated.
For each patient the depth and thickness associated to plane angioma in the specific region
under treatment evaluated with the PT model has been showed. The depth estimated mainly
changes in the range from 300 μm to 700 μm while the thickess changes from 100 μm to 300
μm. Moreover the deeper and thicker plane angioma has been found in the cheek region as an
average, while the less depth and less thick has been found on the temple region.
These evaluations allowed a specific morphological reconstruction of the multilayer biological
tissue under treatment for each clinical situation. On these multilayer the 2D+1 numerical
simulation has been executed. The numerical program of calculation has been realized with a
“home-made” Matlab code.
123
In the simulations the thermal behaviour of the biological multilayers has been achieved for
different set of laser parameters as input. A typical case of temporal sequence made of grid
representing the temperature distribution and calculated with the numerical model has been
shown and discussed.
Next, numerical solutions refered to a same structure and for different laser parameters have
been illustrated. The temporal evolutions and the spatial distribution of the temperature on the
biological layers have been compared. The induced thermal gap ( T) calculated between the
plane angioma and epidermis at istant t=0 s refering to the reference laser parameters has been
found ~ 15 °C. Changing the spot diameter or the pulse duration the induced T can increase
up to ~ 25 °C and up to ~ 45 °C respectively.
At a second stage, numerical solutions refering to different structures with the same laser
parameters have been compared. In a first case, the results refering to three structures with a
plane of blood vessels at different depth (900 μm, 600 μm and 300 μm) but with the same
thickess (100 μm) have been reported. In this situations, T moves from ~23 °C to ~ 40 °C
(with a gap of ~17°C) for epidermis, and moves from ~25 °C to ~75 °C (with a gap of ~50°C)
for plane angioma layers, when deeper angioma are taken. In a second case, two biological
structures with a plane angioma at same depth (600 μm) but with different thickness (100 μm
and 300 μm) have been compared. Differently from the first case, the variations on the
temperatures achieved for the plane angioma and for epidermis has been, substantially, the
same. In fact, the maximum T found for the plane angioma goes from ~42 °C to ~52 °C while
the maximum T found for the epidermis goes from ~ 28 °C to ~37 °C, with a gap in both
situations of about ~9-10°C.
These results demonstrate how the thermal behaviour achieved under treatment, on the
multilayer biologial structures affected by plane angioma, is more sensitive to the variations of
the depth than variations of the thickness.
Subsequently, the thermal effects achieved on different phototypes have been compared. This
last evaluation emphasizes how the thermal gap between epidermal layer and plane angioma
layer decreases going from phototype I to phototype VI. In particular, for phototype VI the
laser pulse induces a maximum T on the epidermis (for t= 0 s) higher than the maximum T
obtained on the angioma.
In all situations the experimental values and the numerical results have been compared and, in
the majority of cases, a good agreement has been obtained for both recovery time and temporal
evolution of T, above all, for more superficial angioma.
124
After these results, the limit conditions for application of laser treatments have been discussed,
focusing on particular situation (higher phototype or deeper angioma) where the results
achieved advise against the application of the laser treatments.
From the evaluation realized, it is possible to choose a set of laser parameters for each different
biological multilayer in order to optimize the thermal interaction and so to make laser
treatments safer and more efficient. Some set of laser parameters have been proposed and
illustrated for treatment of patients affected by plane angioma of different dimensions.

In conclusion in this thesis it has been demonstrated that functional IR imaging can be used
with remarkable advantages in monitoring the physical action of the laser ligth thus giving
to the operator (the physician) information to optimize his activity, in terms of both therapy
effect and safety.

Our further aim is to obtain a complete validation of this novel approach on this research
activity will be done according to the next steps:

- Testing new PT model in order to achieve a more accurate estimation of the
dimensional parameters for the plane angioma.
- Gradual application of the proposed laser parameters set on different targets.
- Validation of the operative protocol proposed.
- Prototipal realization of a infrared imaging system to support the medical laser tool.
- Testing this approach for different patologies and in different medical fields.

The output of this research could supply the doctor with very clear indications, according to the
patient, on how to use the laser in terms of power and duration of the treatment. The final
output will be a methodic scheme which could be employed by the operator no longer in an
empirical way but now based on specific patient and phototype skin. The monitoring system
which will be designed in the future in these activities could be of interest for medical
industries which operate in this sector on a national and international scale.




125
References




[1] R.R. Anderson, J.A. Parrish, “The optics of the human skin”, J. Invest Dermatol 77,13 (1981).

[2] M.H. Niemz, “Laser-Tissue Interactions“, Springer Berlin Heidelberg New York, ch. 1 (2003).

[3] S.O. Kasap, “Optoelectronics and Photonics: Principles and Practices”, Prentice-Hall, ch.1 (2001).

[4] A. Yariv, “Quantum Electronics”, John Wiley Interscience, ch. 1 (1975).

[5] L. Henvey, J. Greenstein, “Diffuse radiation in the galaxy”, Astrophys. J. 93, pp. 70-83 (1941).

[6] H.C. van de Hulst, “A new look at multiple scattering”, Tech. Report NASA, New York, (1962).

[7] J.H. Joseph, W.J. Wiscombe, J.A. Weinman, “The delta-Eddington approximation for radiative
flux transfer”, J. Atmos. Sci. 33, pp. 2452-2459 (1976).

[8] L.O. Reynolds, N.J. McCormick, “Approximate two parameter phase function for light scattering”,
J. Opt. Soc. 70, pp. 1206-1212 (1980).

[9] A. Vassiliadis, H.C. Christian, K.G. Dedrick, “Ocular laser threshold investigations”, Aerospace
Med. Rep. F41609-70-C0002 (1971).

[10] W.S. Weinberg, R. Birngruber, B. Lorenz, “The change in light reflection of the retina during
therapeutic laser-photocoagulation”, IEEE J. Qu. Electron. QE – 20, pp. 1481-1489 (1984).

[11] A.N. Takata, L. Zaneveld, W. Richter, “Laser induced thermal damage in skin. Aerospace Med”,
Rep. SAM-TR- pp. 77-38 (1977).

[12] F.C. Henriques, “Studies of thermal injury”, Am. J. Pathol. 23, pp. 489-502 (1947).

[13] A. Roggan, G. Muller, “Computer simulations for the irradiating planning of LITT”, Med. Tech. 4,
pp. 18-24 (1993).

[14] P. Teng, N.S. Nishioka, R.R. Anderson, T.F. Deutsch, “Acoustic studies of the role of immersion
in plasma-mediated laser ablation”, IEEE J. Qu. Electron. QE-23, pp. 1845-1852 (1987).

[15] D. Stern, R.W. Schoenlein, C.A. Puliafito, E.T. Dobi, R. Birngruber, J.G. Fujimoto, “Corneal
ablation by nanosecond, picosecond and femtosecond laser at 532 nm and 625 nm”, Arch.
Ophthalmol. 107, pp. 587-592 (1989).

[16] M.H. Niemz, E.G. Klancnik, J.F. Bille, “Plasma mediated ablation of corneal tissue at 1053 nm
using a Nd:YLF oscillator/regenerative mplifier laser”, Lasers Surg. Med. 11, pp. 426-431 (1991).

[17] F. Seitz, “On the theory of electron multiplications in crystals”, Phys. Rev. 76, pp. 1376-93 (1949).

[18] A.G. Molchanov, “Avalanche ionization in transparent dielectrics induced by intense light pulses”,
Sov. Phys. Solid State 12, 749, (1970).

126
[19] N. Yablonovitch, N. Bloembergen, “Avalanche ionization and the limiting parameter of filaments
induced by light pulses in transparent media”, Phys. Rev. Lett. 29, pp. 907-910 (1972).

[20] N. Bloembergen, “Laser induced electric breakdown in solids”, IEEE J. Qu Electron. QE-10,
pp. 375-386 (1974).

[21] A.S. Epifanov, “Theory of electron-avalanche ionization induced in solids by electromagnetic
waves”, IEEE J. Qu Electron. QE-17, pp. 2018-2022 (1981).

[22] C.A. Sacchi, “Laser-induced electric breakdown in water”, J. Opt. Soc. Am. B8, pp. 337-45 (1991).

[23] C.A. Puliafito, R.F. Steinert, “Short-pulsed Nd:YAG laser microsurgery of the eye: biophysical
considerations”, IEEE J. Qu. Electron. QE-20, pp. 1442-1448 (1984).

[24] M.M. Krasnov, “Laserpuncture of anterior chamber angle in glaucoma”, Am. J. Ophthalmol. 75,
pp. 674-678 (1973).

[25] D. Aron Rosa, J. Aron, J. Griesemann, R. Thyzel, “Use of the Nd:YAG laser to open the posterior
capsule after lens implant surgery: a preliminary report”, J. Am. Intraocul. Implany. Soc. 6,
pp. 352-354 (1980).

[26] F. Frankhauser, P. Roussel, J. Steffen, E. Van der Zypen, A. Chrenkova, “Clinical studies on the
efficiency of high power laser radiation upon some structures of the anterior segment of the eye”,
Int. Ophthalmol. 3, pp. 129-139 (1981).

[27] M.J.C. van Gemert, A.J. Welch, “Time costants in thermal laser medicine”, Laser Surg. Med. 9,
pp. 405-421 (1989).

[28] W. Seipp, D. Haina, V. Seipp, “Laser in Dermatologie. Ing. Angewandte Lasermedizin“, eds. H.P.
Berlien, G. Muller, Ecomed – Verlag, Landsberg (1989).

[29] O.T. Tan, K. Sherwood, B.A. Gilchrest, “Treatment of childen with port-wine stains using the
flashlamp-pulsed tunable dye laser”, N. Engl. J. Med. 320, pp. 416-421 (1989).

[30] R.R. Anderson, J. A. Parrish, “Selective photothermolysis: precise microsurgery by selective
absorption of pulsed radiation”, Science 220, pp. 524-527 (1983).

[31] S. Kimel, L.O. Svaasand, T.E Milner, M. Hammer-Wilson, M.J. Shell, J.S. Nelson, M.W. Berns,
“Laser photothermolysis of single blood vessels in the chick chorioallantoic membrane (CAM)”,
Proc. SPIE 2077, pp. 216-226 (1994).

[32] L.G. Astafeva, G.I. Zheltov, A.S. Rubanov, “Modeling of Blood Vessel Heating by Laser
Radiation”, Phys. Qu. Opt. 90, 2, pp. 287–292 (2000).

[33] M.J.C. van Gemert, A.J. Welch, J.W. Pickering, O.T. Tan, G.H. Gijsberg, “Wavelengths for laser
treatment of port wine stains and teleangectasia”, Laser Surg. Med. 16, pp. 147-155 (1995).

[34] A. Scheibner, G. Kenny, W. White, R. G. Wheeland, “A superior method of tatoo removal using
the Q-switched ruby laser”, J. Dermatol. Surg. Oncol. 16, pp. 1091-1098 (1990).

[35] C.R. Taylor, R.W. Gange, J.S. Dover, T.J. Flotte, E. Gonzalez, N. Michauud, R.R. Anderson,
“Treatment of tatoos by Q-switched ruby laser. A dose response study”, Arch. Dermatol. 126,
pp. 893-899 (1990).

127
[36] M.H. Niemz, “Cavity preparation with the Nd :YLF picosecond laser”, J. Dent. Res. 74,
pp. 1194-1199 (1995).

[37] M. H. Niemz, “Ultrashort laser pulse in dentistry- advantages and limitations”, Proc. SPIE 3255,
pp. 84-91 (1998).

[38] W.F. Wieland, H. Nicolai, W. Rössler, F. Hofstädter, Proc. LASERmed 93. Springer-Verlag,
Berlin, Heidelberg, New York (1993)

[39] M.H. Niemz, T.P. Hoppeler, T. Juhasz, J.F. Bille, “Intrastromal ablations for refractive corneal
surgery using picosencd infrared laser pulse“, Laser light Opthalm. 5, pp. 149-155 (1993).

[40] M. H. Niemz, F. H. Loesel, M. Fischer, C. Lappe, J.F. Bille, “Surface ablation of corneal tissue
using UV, green and IR picosecond laser pulse”, Proc. SPIE 2079, pp. 131-139 (1994).

[41] R. Spring, J. Snell, “Infrared thermography”, The Engine Yearbook (2004).

[42] X.P.V. Maldague, “Theory and Practice of Infrared Technology for Nondestructive Testing”,
John Wiley & Sons, New York (2001).

[43] H. Kaplan, “Practical applications of infrared thermal sensing and imaging equipment”,
Proc. SPIE TT13, 137, (1993).

[44] H. Kaplan, “Process control using IR sensors and scanners”, Photonics Spectra, Dec,
pp. 92-95 (1997).

[45] S. Marinetti, D. Robba, F. Cernuschi, P.G. Bison, E. Grinzato, “Thermographic inspection of TBC
coated gas turbine blades: Discrimination between coating over-thicknesses and adhesion
defects”, Infrared Physics & Technology 49, pp. 281-285 (2007).

[46] S.Å. Ljungberg, “Infrared techniques in building and structures: operation and maintenance”, Ed.
X.P.V. Maldague, “Infrared Methodology and Technology, International Advances in
Nondestructive Testing”, Monograph Series, pp. 211-252 (1994).

[47] E. Grinzato, G.P. Bison, S.Marinetti, “Moisture evaluation by dynamic thermography data
modelling”, Thermosense XVI, SPIE 2245, Orlando (USA) (1994).

[48] E. Rosina, N.Ludwig, L. Rosi, “Optimal conditions to detect moisture in ancient buildings. Study
Cases from Northern Italy”, Thermosense XX-An International Conference in thermal sensing
and imaging, diagnostic applications, Orlando (USA) (1998).

[49] E. Grinzato, G.P. Bison, S. Marinetti, V. Vavilov, “Thermal infrared non destructive evaluation of
moisture content in building” CNR Congress on Moisture, Varenna (Italy) (1994).

[50] Q. Li, E. Liasi, R. Du, J. Bujas-Dimitrijevic, A. Chen, “Heating of industrial sewing machine
needles: FEA model and verification using IR radiometry”, eds. D.H. Lemieux, J.R. Snell,
Thermosense XXI, Proc. SPIE, 3700, pp. 347-357 (1999).

[51] S.M. Shepard, “Advances in Pulsed Thermography”, Proceedings of SPIE, Thermosense XXIII,
Orlando (USA), pp. 511–515 (2001).

[52] T. Sakagami, S. Kubo, “Applications of pulse heating thermography and lock-in thermography to
quantitative nondestructive evaluations”, J. Infrared Phys Technol 43, pp. 211–218 (2002).

128
[53] N. Ludwig, P. Teruzzi, “Heat losses and 3D diffusion phenomena for defect sizing procedures in
video pulse thermography”, J. Infrared Phys. Technol. 43, pp. 297–301 (2002).

[54] N.P. Avdelidis, D.P. Almond, “Through skin sensing assessment of aircraft structures using
pulsed thermography”, NDT&E Int. 37, pp. 353–362 (2004).

[55] X.P.V. Maldague, A. Ziadi, M. Klein, “Double pulse infrared thermography”, NDT&E
International 37, pp. 559–564 (2004).

[56] A. Bendada, F. Erchiqui, M. Lamontagne, “Pulsed thermography in the evaluation of an aircraft
composite using 3D thermal quadrupoles and mathematical perturbations”, Inverse Problems 21,
pp. 857–877 (2005).

[57] A. Bendada, C. Ibarra-Castanedo, X.P.V. Maldague, “A combinated integral transform asymptotic
expansion method for the characterization of interface flaws through pulsed infrared
thermography”, QIRT Journal 4, 1, pp. 3-23 (2007).

[58] D. Maillet, J. C. Batsale, A. Bendada, A. De Giovanni, “Méthodes intégrales et contrôle non
destructif par thermographie infrarouge stimulée“, Rev Gen Therm 35, pp. 14-27 (1996).

[59] D.A. Gonzalez, C. Ibarra-Castanedo, X. Maldague, “New algorithm based on the hough transform
for the analysis of pulsed thermographic sequences”, NDT and E Internat. 39, pp. 617-621 (2006).

[60] X.P.V. Maldague, Y. Largoutit, J.P. Couturier, “A study of defect depth using neural networks
in pulsed phase thermography: modelling, noise, experiments”, Rev. Gen. Therm. 37,
pp. 704-717 (1998).

[61] X.P.V. Maldague, S. Marinetti, “Pulse phase infrared thermography”, J. Appl. Phys. 79, 5,
pp. 2694- 2698 (1996).

[62] X.P.V. Maldague, F. Galmiche, A. Ziadi, “Advances in pulsed phase thermography”, J Infrared
Phys. Technol. 43, pp. 175–81 (2002).

[63] P. Cielo, X.P.V. Maldague, A. Deem, R. Lewak, “Thermographic NDE of industrial materials and
structures”, Mater. Eval. 45, 12, pp. 452-460 (1987).

[64] J. Allport, J. McHugh, “Quantitative evaluationof transient video thermography”, eds. D.O.
Thompson and D.E. Chimenti. Review of progress in quantitative nondestructive evaluation, 7A,
pp. 253-262 (1988).

[65] V.P. Vavilov, R. Taylor, “Theoretical and practical aspects of the thermal nondestructive testing
of boned structures”, eds. R.S. Sharp. Research techniques in nondestructive testing, 5,
pp. 238-279 (1982).

[66] H.S. Carlslaw, J.C. Jaeger, “Conduction of Heat in Solids”, 2nd ed. Oxford University (1959).

[67] D. Maillet, A.S. Houlbert, S. Didierjean , A.S. Lamine, A. Degiovanni, “Non-destructive thermal
evaluation of delaminations in a laminate: I. Identification by measurement of thermal contrast”,
Composites Sci. Technol. 47, pp. 137–53 (1993).

[68] A. Bendada, “Reconstruction of a non uniform interface thermal resistance by inverse conduction”,
Inverse Problems in Engineering, pp. 79–123 (1998).

129
[69] D.A. Gonzalez, C. Ibarra-Castanedo, F.J. Madruga, X.P.V. Maldague, “Differentiated absolute
phase contrast algorithm for the analysis of pulsed thermographic sequences”, Infrared Physics
and Technology 48, pp. 16–21 (2006).

[70] D.A. González, C. Ibarra-Castanedo, M. Pilla, M. Klein, J.M. López-Higuera, X.P.V. Maldague,
“Automatic Interpolated Differentiated Absolute Contrast Algorithm for the Analysis of Pulsed
Thermographic Sequence”, Proc. 7th Conference on Quantitative InfraRed Thermography
(QIRT), Rhode Saint Genèse (Belgium), pp. 5-8, H.16.1-H.16.6 (2004).

[71] C. Ibarra-Castanedo , A. Bendada, X.P.V. Maldague, “Image and signal processing techniques in
pulsed thermography”, GESTS Int’l Trans. Computer Science and Engr., 22, 1, pp. 89-100 (2005).

[72] L.D. Favro, H.J Jin, Y.X. Wang, T. Ahmed, X.Wang, P.K. Kuo, R.L. Thomas, “IR Thermal wave
tomographic studies of structural composites”, Thompson and D.E. Chimenti, Review of progress
in quantitative nondestructive evaluation, 11A, pp. 447-452 (1998).

[73] D. Wu, G. Busse, “Lock-in Thermography for NonDestructive Evaluation of Materials”, Rev. Gén.
Therm., 37, pp. 693-703 (1998).

[74] W. Bai, B.S. Wong, “Evaluation of Defects in Composite Plates under Convective Environments
using Lock-In Thermography”, Meas. Sci. Technol. 12, pp. 142-150 (2001).

[75] C. Meola, G.M. Carlomagno, A. Squillace, G. Giorleo, “Nondestructive Control of Industrial
Materials by Means of Lock-In Thermography”, Meas. Sci. Technol. 13, pp. 1583-1590 (2002).

[76] D.L. Balageas, P. Levesque, A. Deem, “Characterization of electromagnetic fields using a lock-in
infrared thermographic system”, ed. L.R. Allen, Thermosense XV, Proceedings SPIE,
pp. 274-285 (1993).

[77] R. Gupta, O. Breitenstein, “Unsteady-state lock-in thermography – Applications to shunts in solar
cells”, QIRT Journal 4, 1, pp. 85-105 (2007).

[78] D. Wu, H. Hamann, A. Salerno, G. Busse, “Lock-in thermography for imaging of modulated flow
in blood vessels”, Proc. QUIRT-96, Eurotherm Seminar 50, pp. 343-347 (1996).

[79] D. Pajani, “Thermographie IR: quelle longueur d’onde choisir I“, Measures, May, 71-76 (1987).

[80] D. Pajani, “Thermographie IR: quelle longueur d’onde choisir II“, Measures, June, 77-80 (1987).

[81] J.T. Woolaway, “New sensor technology for 3 to 5 μm imaging band”, Photonics Spectra. 25, 2,
pp. 113-119 (1991).

[82] A. Merla, L. Di Donato, S. Di Luzio,G. Romani, “Quantifying the Relevance and Stage of Disease
with the Tau image Technique”, IEEE Engineering in Medicine and Biology Magazine 21, 6,
pp. 86-91 (2002).

[83] A. Merla, G. Romani, “Biomedical Applications of Functional Infrared Imaging”, Proc. of the
IEEE Engineering in Medicine and Biology 27th Annual Conference, Shanghai (China) (2005).

[84] A Merla, L Di Donato, G. Romani, “Assistance to surgery by means of infrared functional
imaging: preliminary results”, Proc. of the 23th Annual International Conference of IEEE
Engineering in Medicine and Biology, Proceedings CD, track 1264, Istanbul (Turkey) (2001).

130
[85] A Merla, L. Di Donato, G. Romani, P.Rossini, “Infrared Functional Imaging Evaluation of the
Sympathetic Thermal Response”, Proceedings of the 2nd European Medical and Biological
Engineering Conference EMBEC02, 2, 1610-1611 (2002).

[86] I. Pavlidis, J. Levine, P. Baukol, “Thermal imaging for anxiety detection”, IEEE Workshop on
computer Vision Beyond the Visible spectrum: Methods and Applications. pp. 104-109, Hilton
Head Island, South Carolina (USA) (2000).
[87] M. Garbey, N. Sun, A. Merla, I. Pavlidis, “Contact-Free Measurement of Cardiac Pulse Based on
the Analysis of Thermal Imagery”, IEEE Engineering in Medicine and Biology Magazine, 54, 8,
pp. 1418-1426 (2007).
[88] A. Merla, A. Ledda, A. Di Donato, G. Romani, “Assessment of the effects of the varicocelectomy
on the thermoregulatory control of the scrotum”, Fertility and Sterility, 81, 2, pp. 471-472 (2004).
[89] A. Merla, A. Ledda, L. Di Donato, G. Romani, “Use of Infrared Functional Imaging to detect
impaired thermoregulatory control in men with asymptomatic varicocele”, Fertility and Sterility
78, 1, pp. 199-200 (2002).
[90] A Merla, A Ledda, L Di Donato, G. Romani, “Dynamic Digital Telethermography: A novel
approach to the diagnosis of varicocele”, Medical & Biological Engineering & Computing, 37,
pp. 1080-1081 (1999).
[91] A. Merla, L. Di Donato, G. Romani, P. Rossini, “Recording of the Sympathetic Thermal Response
by means of Infrared Functional Imaging”, Proc. of the 25
th
Annual International Conference of
the IEEE EMBS, Catalog No: 03CH37439C, Cancun (Mexico) (2003).
[92] A. Merla, L. Di Donato, F. Salsano, G. Romani et al., “Raynaud’s Phenomenon: infrared
functional imaging applied to diagnosis and drugs effects”, International Journal of
Immunopathology and Pharmacology 15, 1, pp. 41-52 (2002).
[93] A. Merla, F. Salsano, G. Romani et al., “Infrared Functional Imaging Applied to Raynaud’s
Phenomenon”, IEEE Engineering in Medicine and Biology Magazine 21, 6, pp. 73-79 (2002)
[94] A. Merla, , L Di Donato, G. Farina, S. Pisarri, M. Proietti, F. Salsano, G. Romani, “Study of
Raynaud’s Phenomenon by means of infrared functional imaging”, Proc. of the 23
th
Annual
International Conference of the IEEE EMBS, Istanbul (Turkey) (2001).
[95] O. Schuhfried, G. Vacariu, T. Lang, M. Korpan, H. P. Kiener, V. Fialka-Moser, “Thermographic
Parameters in the Diagnosis of SecondaryRaynaud’s Phenomenon”, Arch Phys. Med. Rehabil.
81 (2000).

[96] M. Garbey, A. Merla, I. Pavlidis, “Estimation of Blood Flow Speed and Vessel Location from
Thermal”, Video. Proc. of the IEEE Computer Society Conference on Computer Vision and
Pattern Recognition (CVPR’04), 1, pp. 356-363 (2004).

[97] N. Sun, M. Garbey A. Merla, I. Pavlidis, “Imaging the Cardiovascular Pulse”, Proc. of the IEEE
Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’05) , 2,
pp. 416-421 (2005).

[98] N. Sun, I. Pavlidis, M. Garbey, J. Fei, “Harvesting the thermal cardiac pulse signal”, In
International Conference on Medical Image Computing and Computer Assisted Intervention
(MICCAI’06), 2, pp. 569 -576 (2006).

131
[99] M. Yang, Q. Liu, T. Turner, Y. Wu, “ Vital sign estimation from passive thermal video”,
Computer Vision and Pattern Recognition. IEEE Conference, pp. 1-8 (2008).

[100] J. R. Keyerserlingk, P. D. Ahlgren, E. Yu, N. Belliveau, M. Yassa, “Functional Infrared Imaging
of the breast”, IEEE Eng. Med. Biol, 19, 3, pp.30-42, (2000).

[101] M. Anbar, C. Brown, L. Milescu, J. Babalola, L. Gentner, “The potential of dynamic area
telethermometry in assessing breast cancer”, IEEE Eng. Med. Biol, 19, 3, pp.58-62 (2000).

[102] M. Anbar, C. Brown, L. Milescu, J.A. Babalola, ”Clinical Applications of DAT using a QWIP
FPA camera”, SPIE Proc. 3698, pp. 93-104. (1999).

[103] S.W. Lanigan, J.A. Cotterill, “ Psychological disabilities amongst patients with port wine
stains”, Br. J. Dermatol. 121, 2, pp. 209-15 (1989).

[104] A.W. Bush A, ”Perturbation Methods for Engineers and Scientists”, Boca Raton, FL:CRC (1992).


[105] A. Aziz , T. Na , “Perturbations Methods in Heat Transfer”, Berlin: Springer (1984).


[106] A. Bendada, “Approximate solutions to three-dimensional unsteady heat conduction through
plane flaws within anisotropic media using a perturbation method”, Modelling Simul. Mater.
Sci. Eng. 10, pp. 673–84 (2002).

[107] D. Maillet, S. Andre, J.C. Batsale, A. Degiovanni, C. Moyne, “Thermal Quadrupoles: Solving
the Heat Equation Through Integral Transforms”, Chichester: Wiley (2000)

[108] H. Stehfest, “Remarks on algorithm 368, numerical inversion of Laplace transforms”, Commun.
ACM,pp. 624, (1970).


[109] D. Maillet, A. Houlbert, S. Didierjean , A.S. Lamine, A. Degiovanni, “Non-destructive thermal
evaluation of delaminations in a laminate: I. Identification by measurement of thermal contrast”,
Composites Sci. Technol. 47, pp. 137–53 (1993).


[110] J. Werner, M. Buse, “Temperature Profiles with Respect to Inhomogeneity and Geometry of the
Human Body”, Journal of Applied Physiology 65, 3, pp. 1110-1118 (1988).


[111] L. Myron, P. Cohen, “Measureament of the thermal properties of the human skin. A review”, The
J. Invest. Dermatol. 69, pp. 333-338 (1977).

[112] A.J. Welch, M.J.C. van Gemert, “Optical-Thermal Response of Laser-Irradiated Tissue”, ed.
Plenum Press, NY (1995).

[113] L.F. Douven, G.W. Lucassen, “ Retrieval of opticalproperties of skin from measurement and
modelingthe diffuse reflectance“, Proc. SPIE, 3914, pp. .312-323 (2000).

[114] W.G. Zijlstra, A. Buursma, O. W. van Assendelft, “Visible and Near Infrared Absorption
Spectra of Human and Animal Haemoglobin”, VSP Publishing, Utrecht (2000).


132
[115] A. Roggan, M. Friebl, A. Hahn, G. Muller, “Optical properties of circulating human blood in the
wavelength range 400-2500 nm“, J. Biomed. Opt. 4, pp.36-46, (1999).

[116] J. W. Pickeringt, P.H. Butler, B.J. Ring, E.P. Walker, “Computed temperature distributions
around ectatic capillaries exposed to yellow (578 nm) laser light”, Phys. Med. Biol. 34, 9,
pp. 1247-1258 (1989).

[117] G.W. Lucassen, W.V., M. Keijzer, M.J.C. van Gemert, “Light Distributions in a Port Wine Stain
Model Containing Multiple Cylindrical and Curved Blood Vessels”, Lasers in Surgery and
Medicine 18, 345-357 (1996)

For Veronica, the light of my life

TABLE OF CONTENTS

Introduction......................……………………........................................................…............. i-iv

Chapter 1
LASER – BIOLOGICAL TISSUE INTERACTION

1.1 1.2 1.3 1.4 1.5 1.6

Introduction……........................................................................................ Reflection and refraction of the tissue....................................................... Absorption of the biological tissue............................................................ Scattering................................................................................................... Turbid media............................................................................................................ Laser-Tissue: Interaction Mechanisms...................................................... 1.6.1 Thermal Interaction......................................................................... 1.6.2 Photochemical Interaction............................................................... 1.6.3 Photoablation……………………………………………………... 1.6.4 Plasma-induced Ablation……………………………………….... 1.6.5 Photodisruption…………………………………………………... 1.7 Laser Applications in the medical field………………………………….

1 2 3 5 7 9 10 19 20 23 25 27

Chapter 2
INFRARED IMAGING: Principles and applications

2.1 Introduction………... ……………………………………………………. 2.2 Passive thermography…………………………………………………..... 2.3 Active thermography: Testing procedures…………………………......... 2.3.1 Pulsed Thermography…………………………………………..... 2.3.1.1 The complete thermogram sequence…………………… 2.3.1.2 Definitions of thermal contrast………………………..... 2.3.2 Step Heating (Long Pulse)……………………………………...... 2.3.3 Lock-in Thermography…………………………………………… 2.3.4 Vibrothermography…………………………………………….… 2.4 Infrared detectors………………………………………………….……... 2.4.1 Atmospheric Trasmittance……………………………………….. 2.4.2 Bolometers……………………………………………………...… 2.4.3 Pyroelectric detectors…………………………………………..… 2.4.4 Quantum Detectors……………………………………………….. 2.5 Infrared Imaging Devices: Focal Plane Arrays (FPA)…………………… 2.5.1 Detectors: Performance parameters………………………………. 2.6 New approach of infrared imaging in medical science………………….

32 34 35 37 40 42 47 49 50 52 53 55 56 57 58 60 62

..……………… 120 References..……………………………………….... 3... Infrared monitoring of plane angioma treatment: approach description…...............4.....……......... Numerical simulations: results…………………….........….....….1 3......…...……………......…………………....………...1 4..1 Formulation of the forward problem: 3D thermal quadrupoles...……………… 125 ........... 88 89 92 97 100 116 117 General Conclusions....1 Model Computation…………………..…....5 4..……..................…………........………. Plane angioma pathology (Port Wine Stain)………………………….2 3..5................……....……………........7 Introduction………………………………………………..…………………….....3 4......... Final conclusions: set of laser parameters….......4. 3.............3 Evaluation of the depth……………………………………………..…...........4 4...5 Statistical analysis of the accuracy of the inversion procedures…… 3........2 4...5 Numerical simulation: biometric heat transfer model..…......... 3.... Preliminary results and observations....………………………………..………...... 3...…………...6 4.2 Resolution of the forward problem: mathematical perturbations…......3 3........4......4 Evaluation of the thickness....... Pulsed Thermography model: experimental results ......………......4 65 65 70 72 73 77 80 81 82 84 86 Chapter 4 EXPERIMENTAL AND NUMERICAL RESULTS 4........4... Pulsed Thermography model……………………………………………… 3.......….................4....Chapter 3 INFRARED MONITORING ON PLANE ANGIOMA: PT model and numerical simulation Introduction…………………………………………………………….. Limit conditions for applications of laser treatments……. 3.. Materials and method…... 3.........

The activity of the doctor consists in applying the laser light for a specific time and with a fixed intensity in the area of interest of the skin of the patient. Nevertheless nobody could have imagined at that time the development and the enormous effect of its use in so many different sectors of medicine ranging from surgery to oncology. In most of the treatments what happens is that the doctor has no possibility to directly evaluate the effect of the laser applied on line since the wavelength of i . From the beginning it was considered as a very potential and promising tool for several different application in medicine. Dermatology is one of the medical field in which laser action has been succesfully employed. and so on. during the last decades. from physiotherapy to aesthetic medicine. Deseases like skin melanoms and angioma are usually treated with laser theraphy. just according to his own experience he decides the duration of the application and the intensity of laser light on the skin according to some parameters such as the individual patient. the pathology and the skin phototype. or to skin tumours treated with coherent light. Nowadays the use of laser has become a very common practice in all the medicine sectors and with different pathologies. In this field. These pathologies have been cured thanks to the use of different laser sources. from ginecology to dermatology. ophtalmology. selected as a function of both wavelengths and power. After. to get up to aesthetic. he verifies the effects induced by the light laser after one or two weeks time. Nowadays dermatologists can rely on a wide range of lasers which vary in wavelengths (some of them out of the visible range) and thus they can be selectively employed according to the specific application and type of pathology. technology supplied laser sources with technical properties specifically designed in order to accomplish the specific requirements of several application. and to more and more sophisticated technologies. We can just think to very complicated surgery operation which are performed in a successful way with the help of laser-scalpel. from urology to ophtalmology.Introduction The application field that attracted a great deal of attention has been the medicine since the birth of the laser. Basically. to operation with laser endoscopy which is now the common practice. and dermatological operations. The physician usually makes use of the laser on his patient in an empirical way.

To enhance the performances of the light effect enlarging remarkably the application fields 2. it is possible to evaluate the effects of such an interaction studying the thermal dynamic and in general the thermal behaviour of the tissues. In the research activity of this work-thesis. It is important to underline that the doctor should always work under a safe regime for the patient. This limits the operation of the doctor because the beam power chosen might be too low or too high on one side or the duration of the pulse too short or too long respectively. And this seems to be the only remarkable limit in the application of laser in dermatology. The information comes only after a few days or weeks later to the physician. In photothermal regime. This technique makes possible the direct evaluation about the reaction of the tissue in the specific region treated. since the effect of the application of laser light on skin is linked to the local temperature induced.To optimize the action in itself in terms of both efficiency and curing times. This means that he has to avoid damages like scars which might occur since he has no direct vision of the interaction light-treated tissues and so he usually prefers working with a lower intensity of light. it is possible to achieve information about thermal interaction of surface and subsurface structure of the target under investigation by analysis and images elaboration. The equipment capable of performing such measures is the infrared camera since it supplies thermal images which represent the dynamic thermal evolution laser induced. the main pathology monitored and studied by infrared technique has been the Plane Angioma known as Port Wine Stain (PWS) pathology. This means flashes of few milliseconds too fast to be recorded by the eyes. ii . This pathology is a vascular malformation that consists of a blood vessels accumulation under the tissue. in general. This aspect.it is not in the visible range or it is pulsed. In the laser therapy field. Thanks to this technique. In this view infrared thermography seems to be the best solution. The possibility to give to the doctor a tool which makes him able to monitor on-line the reaction of the skin during the laser application (and not just examining the post application effects) means: 1. on one side means a safer situation for the patient but on the other side minimize the efficacy of the therapy making the cure time longer. infrared imaging technology can be used for a real-time monitoring of the laser-biological tissue interaction. different from case to case. This accumulation form a subsurface plane of vessels of depth and thickness depending on the specific anatomic region and.

in the dermtological field. By the evaluations obtained with this approach. iii .It’s possible to obtain information about the specific plane of blood vessels under treatment monitoring the thermal reaction of the tissue under laser action by IR imaging technique. as a tool for scientific investigation And in particular way . the operator can adjust the laser parameters for each specific therapy in a proper way changing the intensity and the duration of the laser ligth according to the type of plane of vessels and to the skin phototype treated. it is possible to get a local evaluation of the depth and thickness of the plane of blood vessels in the anatomic area that the physician wants to treat. present in the literature. well controlled and less empiric way. With the simulations the thermal behaviour of the biological multilayers can be achieved for different set of laser parameters as input. a 2D+1 numerical simulation based on a biometric heat transfer model is executed on this multilayer. This allows to specialists to employ laser systems in a more specific. The research activity. Subsequently.To evaluate the support of the infrared imaging. object of this thesis work. With this model from the temporal evolution of the temperature obtained on the tissue after a single laser pulse-test (heat souce). In particular.To evaluate the employment of this technique as a tool for the control and optimization of laser parameters in order to make the treatment safer and more efficient In this work a novel methodological approach for the laser parameters optimization in the treatment on plane angioma is proposed. This information permits a morphological reconstruction of the multilayer biological tissue under treatment. an active thermography model. with this approach. is employed in order to find morphological information for a specific plane angioma. The main target of this study are: . theoretical and experimental. is described in four chapters.

the pulsed thermography model and the numerical approach employed in the operative protocol performed are explained in detail. iv . At the end of the chapter. support and courtesy. In particular. the photochemical interaction. is the reading key for the understanding of the experimental work presented in this thesis. in particular. the plasma-induced ablation and photodisruption are described. The data analysis and imaging elaboration have been done in the LIRT of the Cibernetic Institute of CNR “E. The problems relative to atmospheric trasmittance and infrared detectors are treated. The third chapter. The passive and active approach are explained. The experimental research activity has been made in collaboration with the Department of Dermatology of the University II Policlinico in Naples. some set of laser parameters optimized for treatment of specific clinical situations are proposed. In addition I also want to thank Dr. Monfregola of the Department of Dermatology of the University II Policlinico for their availability. our novel approach and . The chapter conclude with the state of art of application in medical field. the pathology analized and computation instruments used are treated and discussed. P. In the conclusions of the chapter. the photoablation. monitoring patients in the laser therpay and photodynamics ambulatory. Numerical and experimental results achieved with the novel approach proposed and relative to different multilayer biological tissues affected by plane angioma are illustrated and compared. the new application of infrared imaging in medical science are discussed. the the laser In the second chapter the priciples and applications of the infrared imaging technique are reported. Caianiello” in Pozzuoli (Na). In the last chapter. the main interaction mechanisms as the photothermal interaction. G. A. The main characteristic of the laser tool and the infrared cameras used in the monitoring are showed. Particularly. Petti to whom go my most sincere thanks. the main experimental and numerical results found are reported and discussed. L. Baldo and Dr. which funded my grant. A detailed description of the activity research is given. The novel approach proposed and realized.In the first chapter the foundamental theoretical concept of the laser–tissue interaction are presented. Some preliminaries results are presented. under the supervision of Dr. Mormile and in collaboration with Dr.

Chapter 1 LASER – BIOLOGICAL TISSUE INTERACTION 1. absorption or scattering. Reflection Incident radiation Absorption and diffused radiation Fig 1. in principle. The optical parameters as the refractive index. When a light beam is incident on a slice of matter. we can have a predominance of either losses or reflection. difficult to measure the effect of refraction due to absorption and scattering. refraction. usually.1 Introduction In this chapter. the absorption and scattering coefficient depend on the wavelength of the incident light. it is. it will be discussed basic phenomena occurring when matter is exposed to light. the index of refractive strongly depends on wavelength in spectral regions of high absorption for the tissue only. Matter can act on electromagnetic radiation in manifold ways. absorption and diffused radiation 1 . refraction plays a significant role only when irradiating transparent media (corneal tissue). three effects exist which may modify its propagation [1-4]: - reflection and refraction absorption scattering In medical laser applications.1 : Geometry of reflection. In particular. however. According to the type of material and the incident wavelenght. In opaque media.

2 Reflection and refraction of the tissue In laser surgery.2 Fig.1. In reality. In general. limitations are given by the inhomogeneity of biological tissue which are also responsible for our inability to provide more than mean parameters of the tissues. 1. n1 and n2). The mathematical relation governing refraction is known as Snell’s law : n1 sin 1  n2 sin  2 When sin 1  n2 / n1 the relation (1. since none of them is a highly polished surface such as an optical mirror. of course. 1. a reflecting surface is the physical boundary between two materials of different indices of refraction such as air and tissue. when the roughness of the reflecting surface is comparable or even larger than the wavelenght of radiation. diffuse reflection occurs.2: Geometry of light propagation 2 . The simple law of reflection states that the reflection angle ` equals the angle of incidence . expressed by:  ' 1. Refraction occurs when the reflecting surface separates two media of different indices of refraction (i.2) is not fulfilled and a total reflection occurs. knowledge of absorbing and scattering properties of a selected tissue is essential in order of predicting a successful treatment. Reflection is defined as the returning of electromagnetic radiation by surfaces upon which it is incident. Diffuse reflection is a common phenomenon of all tissues .e.1 In contrast.

3 Absorption of the biological tissue Absorption is due to a partial conversion of light energy into heat motion of vibrations of molecules of the absorbing material. Whereas absorption in the IR region of the spectrum can be primarily attributed to water molecules. absorption spectra of two elementary biological absorbers are shown. absorption is mainly caused by either water molecules or macromolecules such as proteins and pigments. It has relative absorption peaks around 3 . respectively. In contrast.3 where z denotes the optical axis. The inverse of the absorption coefficient also referred to as the absorption length L is L=1/ 1. In biological tissues. In fig 1. Its absorption coefficient monotonically increases across the visible spectrum toward the UV. The law that descibes the effect of thickness on absorption is called Lambert-Beer law and is expressed by: I ( z )  I 0 exp( z ) 1. I(z) is the intensity at a distance z. is the absorption coefficient of the medium.1. Among biological tissue cornea and lens can be considered as being highly transparent for visible light. Melanin is the basic pigment of skin and is by far the most important epidermal chromophore. They belong to melanin and hemoglobin (HbO2). proteins as well as pigments mainly absorb in the UV and visible range of the spectrum.4 The absorption length measures the distance z in which the intensity I(z) has dropped to 1/e of its incident value I0.3. The terms “transparent” and “opaque” are relative. since they are wavelength dependent. I0 is the incident intensity. Haemoglobin is predominant in vascularized tissues. media in which incident radiation is reduced prectically to zero are called opaque. no tissue is known to be either transparent or opaque to all wavelenghts of the electromagnetic spectrum. A perfectly transparent medium permits the passage of light without any absorption. Actually.

3: Absorption spectra of melanin in skin and haemoglobin (HbO2) in blood 4 . a therapeutic window for many patologies is delineated between roughly 600 nm and 1200 nm. thus enabling treatment of deeper tissue structures. Since the green and yellow wavelengths of Argon and Krypton ion laser perfectly match the absorption peaks of haemoglobin. Fig. In this range. 540 nm and 580 nm. 1. For some clinical applications. Since neither macromolecules nor water strongly absorb in the near IR .280 nm. these lasers can be used for the coagulation of blood and blood vessels. 420 nm. and then exhibits a cut-off at approximately 600 nm. dye lasers may be an alternative choice since their tunability can be advantageously used to match particular absorption bands of specific proteins and pigments (selective absorption) . radiation penetrates biological tissues at lower loss.

Scattering can be elastic or inelastic. Its only restriction is that the scattering particles are smaller than the wavelenght of the incident radiation.1. depending on whether part of the incident photon energy is converted during the process of scattering. In particular. A special kind of elastic scattering is Rayleigh scattering. In paricular. it takes place when charged particles are exposed to electromagnetic waves of frequencies not corresponding to those natural frequency of free vibration of the same particles (resonance frequencies). 1.4 Scattering Scattering is a class of phenomena by which particles are deflected by collisions with other particles.4 Fig.5 If the scattering angle is taken into account. Rayleigh’s law is illustrated in fig.4: Rayleigh’s law of scattering in three different regions 5 . a more detailed analysis yelds IS  1  cos( ) 4 1. This statement is also known as Rayleigh’s law and is given by: IS  (n  1) 2 4 1. it’s possible to find a relationship between scattered intensity and refraction index. Elastic scattering occurs when incident and scattered photons have the same energy. 1.6 where =0 denotes forward scattering.

the coefficient of anisotropy g represents the avrerage value of the cosine of the scattering angle . Therefore. Otherwise.7 to 0. Therefore the relations given are valid only for wavelengths far away from any absorption band. it is very convenient to define a probability function p( ) of a photon to be scattered by an angle which can be fitted to the experimental data. Mie scattering shows a weaker dependeance on wavelenght (~ -x -4 with 0. However. the observed wavelength dependence is somewhat stronger than that predicted by Mie scattering. As a good approximation. A measure of the anisotropy of scattering is given by the coefficient of anisotropy g. Mie scattering preferably takes place in the forward direction. and Parsa et al. When p( ) does not depend on . forward and backward scattered intensities are the same. The theory of Mie scattering is rather complex and will not be repeated here. In the derivation of Rayleigh’law. If the size of the scattering particles becomes comparable to the wavelenght of the incident radiation. the scattering is isotropic.7 where p( ) is a probability function and d  sin( )dd is the elementary solid angle. and g = 0 isotropic scattering. where g = 1 denotes purely forward scattering.6). it was found by Wilson and Adam (1983).e. the coefficient g is defined by g   p( ) cos( )d   p( )d 4 4 1.99 for most biological tissues. anisotropic scattering occurs. neither Rayleigh scattering nor Mie scattering completely describe scattering in tissues. (1. In polar coordinates. (1989) that photons are preferably scattered in the forward direction. Thus. it can be assumed that g ranges from 0.6).5) and (1. absorption has been neglected. This phenomenon cannot be explained by Rayleigh scattering. Second. Jacques et al.4 x 0. Hence the corresponding scattering angles are most frequently between 8° and 45°. First. whereas Rayleigh scattering is proportional to 1+cos2( ) according to (1.Within the visible spectrum. such as in the blood cells. Rayleigh scattering no longer applies and another type of scattering called Mie scattering occurs. scattering is significantly reduced when we shift from blu to red light. By definition . On the other hand. In most biological tissues. 6 . g = -1 purely backward scattering. (1987). it is well known that Mie scattering and Rayleigh scattering differ in two important aspects.5) compared to Rayleigh scattering (~ ). i.

we have considered the occurrences of either absorption or scattering. 1.7 is the probability function p( ). In most tissue both of them are present simultaneously. by a s t 1. More detailied information on these phase functions are provided in the references [5-8]. but it is important to realize the existence of both processes and the fact that usually both are operating.5 Turbid media In the previous two sections. the mean free optical path of incident photons is thus determined by Lt  1 t  1  s 1. = s. whereas in the case of a  1 only scattering occurs. attenuation is exclusively due to absorption. the relation (1.10 For a  0 . also called phase function.The important term in 1.9 Only in some cases either or s may be negligible with respect to each other. For a  0.8 In turbid media. In 7 . both effects will take place but they will occur in variable ratios. Such media are called turbid media. It is very convenient to define an additional parameter.5 .10) can be turned into the equality general. Several theoretical phase function p( ) have been proposed in literature in order to find the best accordance with experimental observation. the optical albedo a . Their total attenuation coefficient can be expressed by t     s 1.

would not lead to an attenuation of intensity.e. Three different absorption coefficient are assumed which are typical of biological tissue.12 since forward scattering alone.5: Optical albedo as a function of scattering coefficient In literature. s >> . i. reduced scattering and attenuation coefficient are often used and are given by  s '   s (1  g ) 1. For albedo asymptotically approaches unity. 1.In fig. 1.4 it is shown the albedo as a function of the scattering coefficient.11 and t '     s ' 1. 8 . g =1. the Fig.

All together. they determine the total trasmission of the tissue at certain wavelengths.1. the parameter that ditinguishes and controls these processes is the exposure time. 1. Although the number of possible combinations for the experimental parmeters is unlimited. Among these. plasma-induced ablation and photodisruption. On the other hand. According to this map. 1. the following parameters are given by the laser radiation itself: wavelenght. photoablation. On the other hand. exposure time. absorption. A map with the five basic interaction types is shown in fig. scattering and others such as heat conduction and heat capacity.6 Fig. five main categories of interaction types are classified today : photochemical interaction. Most important among optical tissue properties are the coefficients of reflection. thermal interaction.6: Map of laser tissue interaction The y axis expresses the applied power density in W/cm2 and the exposure time in seconds is reported on the x axis. energy density and spot size. the time scale can be roughly divided into five 9 . Specific laser characteristics as well as laser parameters contribute to this diversity. the exposure time is a very crucial parameter when selecting a certain type of interaction.6 Laser-Tissue: Interaction Mechanisms The interaction mechanisms that may occur when applying laser light to biological tissue are manifold. All these different interaction types share a single common issue: the characteristic energy density ranges from approximately 1 to 1000 J/cm2 .

the exposure time appears to be the main parameter responsible for the variety of interaction mechanisms. and coagulation tissue becomes necrotic. The resulting ablation is called thermal decomposition . in fact. carbonization should be avoided in any case. In the next sections each of these interactions mechanisms will be briefly discussed whereas a greatest attention will be given to the description of the thermal interaction mechanism which is core of the present thesis work. For instance. vaporization. Anyway.1 Thermal Interaction The thermal interaction term stands for a large group of interaction types. where the increase in local temperature induces the significant parameter change. Thermal effects can be induced by either CW or pulsed laser radiation. Thus. carbonization only reduces visibility during surgery. from 1 μs 1 ns for photoablation.94 μm). thermal effects may also play an important role during photochemical interaction. If a laser with one of these wavelengths is used (i. The difference between the latter two is attributed to different energy densities.e. 2. and melting may be distinguished : . temperature reaches at least 60°C.Carbonization: At temperature over 100°C. carbonization. since tissue already becomes necrotic at lower temperatures.e. 1. strongly absorbs to different wavelength in the NIR region (i. and less than 1 ns for plasma-induced ablation and photodisruption. Thus. the tissue starts to carbonize. . 10 . Depending on the duration and peak value of the tissue temperature achieved. different effects like coagulation. This effect is attributed to the rich presence of water molecules in the tissues.Vaporization: It is a thermomechanical effect due to the pressure build-up involved.sections: continuos wave or exposure times more than 1 s for photochemical interaction. Er:YAG) a higher pressure in the tissue is induced (water tries to expand in volume as it vaporizes) and this produce localized microexplosions. The reciprocal correlation between power density and exposure time clearly demonstrates that roughly the same energy density is required for any type of interaction.6.Coagulation: during the process of coagulation. For medical laser applications. adjacent interaction types present in the map cannot always be strictly separated. Water. from 1 s to 1 μs for thermal interaction.

is the governing parameter of all thermal laser-tissue interactions. First.7: Absorption of water 11 . The absorption coefficient strongly depends on the wavelenght of the incident laser radiation. proteins. inelastic collision with some partner of the surrounding medium lead to a deactivation of A* and a simultaneous increase in the kinetic energy M. and other macromolecules and it is governed by Lambert-Beer law. absorption of a photon with an energy h promotes the molecule to an excited state A*. Temperature. obviously. This two step process can be written as A+ h  A* A* + M(Ekin)  A + M(Ekin + Ekin) - absorption deactivation Both these steps are highly efficient provided that the duration of laser exposure is properly selected. thermal effect have their origin in bulk absorption occuring in molecular vibration-rotation bands followed by nonradiative decay. the temperature rise microscopically originates from the transfer of photon energy to kinetic energy. the absorption spectrum of water. pigments. In biological tissues. absorption is mainly due to the presence of free water molecules. Therefore. The reaction can be considered as a two step process. and second.Melting: They originate from thermal stress induced by a local temperature gradient across the tissue surface. is shown in the next figure Fig. absorption by water molecules plays a significant role. 1. Therefore. an important constituent of most tissues. At the microscopic level. In thermal interactions..

1: Absorption coefficients and absorption length L of water at different wavelength 12 .00029 0.00008 0.0058 0.020 0.61 36 12000 860 10 55 170 430 3400 340 180 50 1.0029 0.1 0. Typical absorption coefficients and the corresponding absorption length L for the most important laser wavelenghts are summarized in table 1. In this last region absorption originates from the presence of resonance frequencies associated to symmetric and asymmetric vibrational modes of water molecules.6 0.57 0.7 1.0056 0. The great attenuation in the UV region is stronghly enhanced by Rayleigh scattering.1.0023 0.028 0.001 Tab 1. Wavelength (nm) Laser type 193 248 308 351 514 633 694 800 1053 1064 2120 2940 10600 ArF KrF XeCl XeF Argon ion He-Ne Ruby Diode Nd:YLF Nd:YAG Ho:YAG Er:YAG CO2 (cm-1) L (cm) 0.As it is possible to see the absorption of water is extremely small in the visible range while in the IR range water molecules are the dominant absorbers.018 0.

t )  I 0 exp(  r2 2  z 2 ) exp(  4t 2 2 ) ) 1. 1.13 2r 2 2  z ) exp(  8t 2 2 where E0 and I0 are the incident values of electric field and intensity. beam waist. and c is the speed of light.8. z. and r the distance from this axis.. is the absorption coefficient.8: Geometry of the tissue irradiation For the sake of simplicity. respectively. the amplitude of the electric field and the corresponding intensity inside the tissue are given by E (r . heat is generated inside the tissue during the laser exposure.14 By means of the two-step process as described above. Fig. t )  E0 exp(  I (r . a cylindrical geometry is chosen with z denoting the optical axis. The incident values E0 and I0 are related to each other by the basic electrodynamic equation 1 2 I 0   0 cE 0 2 where  0 is the dielectic costant . z . We assume that a slab of tissue in air is illuminated by Gaussian-shaped laser beam as illustrated in fig. Deposition of heat in tissue is due only to light that is absorbed in the tissue. 1. and is the is the pulse duration. 1.Heat Generation Heat generation is determined by laser parameters and optical tissue properties. Then. 13 .

t ) z 1. t )  I (r .t).16 Therefore. t )   I (r .8 w )  1. t ) z 1.55  2. z.19 where  is the tissue density and  w is its water content expressed in Kg/m3 14 . z  z. and c is the specific heat capacity. t ) 1. t )  I (r.t) inside the exposed tissue is a function of the absorption coefficient and the local Intensity I(r. z. z. t )  I (r . For most tissues a good approximation is given by c  (1. heat deposition is determined by S (r.z.For a light flux in the z-direction in a nonscattering medium. the last heat source S(r. z. as z approaches zero. z.18 where m is the mass.z. an alteration in heat content dQ induces a linear change in temperature dT according to the basic law of thermodynamics dQ  mcdT 1. we have that S (r .15 And. z. Since is stronghly wavelength-dependent.17 Thus. If phase transitions or tissue alterations do not occur. the local heat deposition per unit area and time in a thickness z is given by S (r . under all circumstances. the same applies for S.

depending on the type of tissue alterations. Thereby. Due to the large increase in volume during this phase transition. since the vapor generated carries away excess heat and helps to prevent any further increase in the temperature of adjacent tissues. As already stated. the fraction of surviving cells is further reduced. melting can occur. Assuming a body temperature of 37°C. To avoid carbonization. Only if all water molecules have been vaporized. where the local temperature and the associated tissue effects are listed 15 . Furthermore. These effects. resulting in reduced energy transfer within the cell and immobility of the cell. At 100°C. a measurable reduction in enzyme activity is observed .Heat Effects Approximate values of achievable temperatures can often estimated. carbonization takes place which is observable by the black-ening of adjacent tissues and the escape of smoke. accompanied by bond destruction and membrane alterations. certain repair mechanism of the cell are disabled. thereby destroying the otherwise maintained equilibrium of chemical concentrations. the tissue is usually cooled down with either water or gas. The first mechanism by which tissue is thermally affected can be attributed to conformational changes of molecules. does the increase in temperature proceeds. The large vaporization heat of water (2253 KJ/Kg) is advantageous. All these steps are summarized in table 1. denaturation of proteins and collagen occurs which leads to coagulation of tissue and necrosis of cells.. The corresponding macroscopic response is the visible paling of the tissue. these can be manifold. and laser exposure is still continuing. a significant percentage of the tissue will undergo necrosis as described below by Arrhenius’ equation. Finally. If such a hyperthermia lasts for several minutes. The thermal interaction dealing with biological effects are related to different temperatures inside the tissue.2. no measurable effects are observed for the next 5°C above this. gas bubbles are formed inducing mechanical ruptures and thermal decomposition of tissue fragments. Beyond 50°C. At even higher temperatures more then 80 °C the membrane permeability is drastically increased. depending on the target material. are summarized in the single term hyperthermia ranging from approximately 42-50 °C. At 60°C. beyond 300°C. water molecules contained in most tissues start to vaporize. At temperatures exceeding 100°C.

Permeabilization of membranes . 1.Protein denaturation 60 .Cell immobility .Temperature (°C) 43 . 1.75 .100 .Formation and breaking of vacuols >300 .9 how the critical temperature and the corresponding temporal duration relate to each other if irreversible damage is meant to occur.Thermoablation of the tissue Tab. 1.9: Critical temperature for the occurrence of cell necrosis 16 .Vaporization 100 . It is illustrated in fig.Conformational alteration .2 : Thermal effects of laser radiation However. Fig. The curve is derived from several empirical observation.45 Biological effect .Collagen denaturation 75 .Hyperthermia .60 .Coagulation .Thermal decomposition (ablation) .Shrinkage of the tissue .Carbonization >100 . it was observed that not only the temperature achieved but also the temporal duration of this temperature plays a significant role for the induction of irreversible damage.Alteration of enzymatic activity 45 .

17 . and the probability of cells staying alive depends on the duration and temporal evolution of the temperature obtained. C(t) is the concentration at a time t. Unfortunately. Areas with maximum temperatures less than 60°C are treated hyperthermically only. depending on the laser parameters. we are able to calculate the probable damage degree Dd(t) as a function of time (t).20 where C0 is the initial concentration of molecules or cells.Areas in which the temperature reaches values higher than 60°C are coagulated and irradiated tissue cells become necrotic. CO2 lasers are very suitable for vaporization and the precise thermal cutting of tissues. However in literature [9-13] some values of various tissues are given as the result of different studies. Carbonization and melting can occur with almost any type of laser if sufficient power densities and exposure durations are provided. not only one but several thermal effects are induced in biological tissues. Laser radiation acts thermally if power densities more than 10 W/cm2 are applied from either CW radiation or pulse durations exceeding approximately 1 μs. and properties. For a quantitative approximation of both the remaining active molecules and cells at certain temperature level. The damage degree is defined as the fraction of deactivated molecules or cells given by E and are specific tissue Dd (t )  C0  C (t )  1  exp( ) C0 1. Typical lasers for coagulation are Nd:YAG lasers. Dye lasers and diode lasers. Frequently. Arrhenius’ equation is very useful t C (t ) E   A exp(  )dt '   0 C0 RT (t ' ) ln 1.21 Thus. by inserting an appropriate value of the tissue constant . A is Arrhenius’constant. experimental data for the two parameters A and E are very difficult to be obtained due to the inhomogeneity of most tissues and the uncertainty in measuring the surviving fraction. These effects might even range from carbonization at the tissue surface to hyperthermia a few millimeters inside the tissue.20) . R is the universal gas constant. The local degree of tissue damage is determined by the damage integral given by (1.

Thus. The location and spatial extent of each thermal effect depend on the locally achieved temperature during and after laser exposure. careful evaluation of the required laser parameters is essential. exposure volume. Therefore. The concomitance of several thermal processes is illustrated in fig. exposure energy. 1.20: Location of thermal effects inside biological tissue 18 .Since the critical temperature for cell necrosis is determined by the exposure time. Fig. no welldefined temperature can be declared which distinguishes reversible from irreversible effects.20. and exposure duration all together determine the degree and extent of tissue damages. 1.

spectrally adaptaded chromophores are injected into the body. highly cytotoxic reactands are released causing an irreversible oxidation of essential cell structures. wavelengths in the visible range (e. The latter are of importance if deeper tissue structures are to be reached.6. In most cases. Rhodamine dye lasers at 630 nm) are used because of their efficiency and their high optical penetration depths.Photodynamic therapy (PDT) During PDT. Careful selection of laser parameters yields a radiation distribution inside the tissue that is determined by scattering. A chromophore compound which is capable of causing light-induced reactions in other non-absorbing molecules is called a photosensitizer. since they were first claimed by the Hungarian surgeon Mester at the end of the 1960s. although this is not scientifically ascertained.Biostimulation Biostimulation is believed to occur at very low irradiances and to belong to the group of photochemical interactions. After resonant excitation by laser irradiation. The potential effects of extremely low laser powers (1-5mW) on biological tissues have been a subject of controversy. the photosensitizer performs several simultaneous or sequential decays which result in intramolecular transfer reactions. so far.1. .2 Photochemical Interaction Photochemical interactions are based on empirical observations stating that light can induce chemical effects and reactions within macromolecules or tissues. biostimulation is also attributed to photochemical interactions. resulting in certain biological transformations. . Frequently. In the field of medical laser physics. photochemical interaction mechanisms play a significant role during photodynamic therapy (PDT). These interactions take place at very low power densities (typically 1 W/cm2) and long exposure times ranging from pulses of seconds to continuous waves. the term biostimulation has not been scientifically very well defined. Wound healing and anti-infiammatory properties by red or near infrared light sources such as helium-neon lasers or diode lasers were 19 . Unfortunately.g. Monochromatic irradiation may then trigger selective photochemical reactions. At the end of these diverse reaction channels.

In the area of such injuries. According to Wilder-smith (1988). conditions are usually create preventing proliferation such as low oxygen concentration or pH. results could be verified by independent research groups. since ln many of these diseases 50 % of the patients are spontaneously cured even without any treatment. However. contradictory results were obtained in many experiments. With this effect a direct breaking of molecular bonds by high-energy (UV) photons is obtained. 1. However. This observation is probably due to the fact that cell proliferation is very active in fresh wounds.reported. The exposure to red or near infrared light might thus serve as a stimulus to increase cell proliferation. The geometry of the ablation pattern itself is defined by the spatial parameters of the laser beam. Detailed investigations in this area and reproducible experimental results are badly needed. its excellent predictability. and the lack of thermal damage to adjacent tissue. the success is rather doubtful. researchers have noticed improvements in the patients. But in a few studies only. According to Karu (1987). biostimulation is a research field with a lot of speculation involved. in which the refractive power of the cornea is altered in myopia. Usually. Furthermore. The main advantages of this ablation technique lie in the precision of the etching process. the distinction from an ordinary “placebo” effect is thus rather difficult to be verified. When irradiating fresh wounds the effect of biostimulation is found to be minimal or even nonexistent. in medical applications. In several cases. 20 . and regeneration is not significantly altered by laser irradiation. only very few patients were treated. photoablation is one of the most successful techniques for refractive corneal surgery. Typical threshold values of this type of interaction are 107-108 W/cm2 at laser pulse durations in the nanosecond range. Moreover.6. Typical energy fluences lie in the range 1-10 J/cm2 . the controversy stems from our inability to specify the photochemical channels of potential reactions. Today.3 Photoablation Photoablation is a laser-tissue interaction where the biological tissue is decomposed when exposed to high intense laser irradiation. and no clinical protocols were established. local wound healing effects with helium-neon or diode lasers may be explained by the action of low-intensity light on o proliferation. The ablation depth (which defines the depth of tissue removal per pulse) is determined by the pulse energy up to a certain saturation limit.

The probability of such transition is increased if the maxima of the corresponding Schrödinger function of the initial and final states coincide at the same radial distance. Thus. Absorption of a photon may promote the two atoms to an excited state (AB)*. let us assume that two atoms A and B are bound by a common electron. It states that the radial distance of the two nuclei of the atoms A and B will not be affected during the process of excitation due to the small electron mass.hyperopia or astigmatism. Thus. the energy gain is usually high enough to access an electronic state which exceeds the bond energy. If a UV photon is absorbed.21: Energy level diagram for photoablation Due to the macromolecular structure.excitation: . 1. the excitation is achieved by satisfying the Franck-Condon principle. each electronic level is split into further vibrational states. In this case. photoablation can be summarised as a two-step process : AB + h  (AB)* (AB)*  A + B + Ek . The corresponding energy level diagram of the ground and several excited states is shown in fig. The character of the bonds within an organic material is primarily covalent.dissociation: 21 . In order to obtain a physical explanation of the photoablation process. Usually.21 Fig. transitions characterized by a vertical line in the energy level diagram are favored. 1. the two atoms A and B may dissociate at the very next vibration.

Rhodamine He–Ne Diode Nd:YAG Er:YAG CO2 193 248 308 351 514 595 633 800 1064 2940 10600 Tab.depending on the absorption coefficient and the incident intensity. This value is reproducible within an uncertainty of about 10% for most materials which is an excellent value when taking all the inhomogeneities of tissues into account. a well-defined depth is ablated .4 2. 1.0 4.2 2. If the incident Intensity is moderate and such that the ablation depth is smaller than the corresponding optical absorption length.0 3.4: Laser systems used in medical applications When comparing both tables. we find that only photons from UV lasers.6 3. Therefore.0 2.1 Dissociation energy (eV) 7. the wavelengths and the corresponding photon energies of selected laser systems are given in the following tables: Laser type Type of bond C=O C=C OH NH CO CC SH CN CS Wavelenght (nm) Photon energy (eV) 6. However. Above this intensity. provide an energy sufficient for dissociating such bonds.5 3.The dissociation energies of some typical chemical bonds.3 : Dissociation energies of some chemical bound Tab. After these a linear relation between the number of applied pulses and the total etch depth is obtained. In practice.4 5. the interactions mechanism of photoablation is limited to the application of UV light.4 4. typically excimer lasers.1 3.2 0. 1. only the first few pulses are unique. subsequent pulses will enter partially irradiated tissues as well as unexposed tissues.6 1.7 ArF KrF XeCl XeF Argon ion Dye .5 2. Therefore.6 3. the etch depths are averaged over several pulses and noted as ablation depth per pulse.4 0.0 1.1 6.8 4. it was observed that a certain threshold intensity must be applied to achieve photoablatlon. 22 .

For picosecond pulses. thus providing the necessary condition for plasma ionization. very clean and well-defined removal of tissue without evidence of thermal or mechanical damage can be achieved. the initial process for the generation of free electrons is supposed to be thermoionic emission.4 Plasma-induced Ablation When obtaining power densities exceeding 1011 W/cm2 in solids and fluids a phenomenon called optical breakdown occurs.22 where  0 is the dielectric constant.6. The electric field strength itself is related to the local power density I by the basic electrodynamic equation 1 2 I 0   0 cE 0 2 1. a very large free electron density with typical value of 1018/cm3 is created in the focal volume of the laser beam. Within a few hundred of picoseconds. typical threshold intensities of optical breakdown are 1011 W/cm2. As result of a optical breakdown a plasma-induced ablation phenomenon can occur. the term multi-photon ionization denotes processes in which coherent absorption of 23 . In mode locked pulses. visible and IR light are strongly absorbed by the plasma. This value is comparable to the average atomic or intramolecular Coulomb electric fields. In Q-swithced pulses. plasma-induced ablation is also referred to as plasmamediated ablation [14-16]. plasma generation due to an intense electric field is called dielectric breakdown. and c is the speed of light. The term optical breakdown especially emphasises that UV. By means of plasma-induced ablation. The most important parameter of plasma-induced ablation is the local electric field strength E which determines when optical breakdown is achieved. If E exceeds a certain threshold value.1. multi-photon ionization may occur due to the high electric field induced by the intense laser pulse. Hence the corresponding electric field amounts to approximately 107 V/cm. The physical principles of breakdown have been investigated in several theoretical studies [17-22]. Plasma-induced ablation permit an ablation of the tissue by ionizing plasma formation. breakdown occurs. In general. if the applied electric field forces the ionization of molecules and atoms. Sometimes. The initiation of plasma generation can be two-fold [23]. In general. It was observed that either Q-swithced pulses in the picosecond or femtosecond range can induce a localized microplasma. for istance.

these free electrons may absorb incoming photons. Each of them will have less individual kinetic energy than the initial electron. release two more electrons. In either case. 24 . Thereby. leading the accumulation of free electrons and ions. The process is schematically represented as : h + e + A+  e + A+ + Ek In particular. multi-photon ionization is achievable only during high peak intensities as in picosecond or femtosecond laser pulses. this event must necessarily take place in the electric field of an ion A+ or a neutral atom. the field of potential medical laser applications is considerably widened. According to Smith and Haught (1966). the important feature of optical breakdown is that it renders possible an energy deposition not only in pigmented tissue but also in nominally weakly absorbing media.like cornea and lens . It is based on a free-free absorption. for istance. however. accelerate. a few “lucky'' electrons initiate an avalanche effect. transparent tissues . Thus.become potential targets of laser surgery. The basic process of photon absorption and electron acceleration taking place in the presence of an atom is called inverse Bremsstrahlung. a transition where a free electron is present in the initial and final states. strike other atoms. In order to achieve optical breakdown the irradiance must be intense enough to cause rapid ionization so that losses do not quench the electron avalanche. Especially in ophthalmology. and so forth. Plasma energies and plasma temperatures are usually higher in Q-swltched laser pulses because of the associated increase in threshold energy of plasma formation. The electron may increase its kinetic energy by means of absorbing any amount of energy. In order to fulfill the conservation laws of energy and momentum. The accelerated electron collides with another atom and ionizes it resulting in two free electrons.several photons provides the energy needed for ionization. optical breakdown achieved with nanosecond pulses is often accompanied by nonionizing side effects. inelastic collisions and diffusion of free electrons from the focal volume are the main loss mechanisms during avalanche ionization. It is also worthwile noting that during the process there is no restriction of the photon energy involved. thus leading to a very short risetime of the free electron density of the order of picoseconds. Again. This is due to the increased absorption coefficient of the induced plasma. Due to the requirement of coherence. A free electron absorbs a photon and accelerates.

the presence of these effects is often an undesired but associated symptom. the tissue is split by mechanical forces. Primarily. As a result of the outer static pressure. the physical effects associated with optical breakdown are plasma formation and shock wave generation. because of the mechanical impact. for nanosecond pulses. By means of the high plasma temperature. because the threshold energy density of optical breakdown is higher compared to picosecond pulses. Whereas plasma-induced ablation is spatially confined to the breakdown region. and the pressure gradient scales with plasma energy. any secondary effects of the plasma have been neglected. this is due to the fact that mechanical effects scale linearly with the absorbed energy. if the bubble is in direct contact with the solid boundary during its collapse.1. the boubble implodes determining a permanent tissue damage. At higher pulse energies. In this case. typically made of water vapor and carbon oxides. Then. If breakdown occurs inside soft tissues or fluid. Thus. picosecond or femtosecond pulses permit the generation of high peak intensities with 25 . thus limiting the localizability of the interaction zone.6. During photodisruption. and thus higher plasma energy. the focal volume is vaporized and a bubble. a high speed liquid jet directed toward the wall is produced. purely plasmainduced ablation is not observed for nanosecond pulses. the jet can cause high-impact pressure against the wall resulting in a severe damage and erosion of tissue. Actually. consequent to jet formation phenomenon. For pulse durations in the nanosecond range. When discussing plasma-induced ablation. the term photodisruption is more appropriate. shock waves and other mechanical side effects become more significant and might even determine the globe effect upon the tissue. Jet formation occur when cavitation bubble collapse in the vicinity of a solid boundary. optical breakdown is always associated with shock wave formation even at the very threshold. Cavitation is an effect that occurs when focusing the laser beam not on the surface of a tissue but into the tissue. bubbles attached to solids have the largest damage potential. Since adjacent tissues can be damaged by disruptive forces. cavitation and jet formation may additionally take place. is formed. Hence. In general. shock wave and cavitation effects propagate into adjacent tissues. the spatial extent of the mechanical effects is already of the order of millimeters even at the very threshold of breakdown.5 Photodisruption Photodisruption effect consist on the fragmentation and cutting of tissues by mechanical forces . In contrast.

This basically is the time needed by the free electrons to diffuse into the surrounding medium. plasma-induced ablation as well as photodisruption.considerably lower pulse energies. Plasma formation begins during the laser pulse and last for a few nanoseconds afterwards. in the 1970s and 1980s. spatial confinement and predictability of the laser-tissue interaction is strongly enhanced. Moreover. the shock wave then 26 . and jet formation . It was only because of recent research that a differentiation between ablations solely due to ionization and ablations owing to mechanical forces seems justified. With these extremely short pulse durations.Percentagesgiven are rough estimates of the approximate energy transferred to each effect The four effects. optical breakdown may still be achieved with significantly reducing plasma energy and. 1. already starts during plasma formation. In general. Since both interaction mechanisms. 1. In fig. Shock wave generation is associated with the expansion of the plasma and. Actually. a schematic sequence of these processes is illustrated indicating their relations to each other.22. cavitation. it is not always easy to distinguish between these two processes. rely on plasma generation. However. disruptive effects. plasma formation. all tissue effects evoked by ultrashort laser pulses were attributed to photodisruption. thus. Moreover. photodisruption may be regarded as a multi-cause mechanical effect starting with optical breakdown. all take place at a different time scale. the distinction between photodisruption and plasma-induced ablation is emphasized Fig. shock wave generation. thus.22: Scheme of the physical process associated with optical breakdown.

the development of new technologies and the increasing understanding of the utilization of that technology by the medical community have brought a continuous increase of laser use in a wide range of medical applications. it has slowed down to on ordinary acoustic wave. Below. After 1974. Nowadays. The high absorption rate in water. hemoglobin. and melanine guarantees efficient clinical applicatios with optimal medical results. finally. lasers have found numerous applications in the field of medicine. since it was introduced by Krasnov (1973) and then further investigated by Aron-Rosa et al (1980) and Fankhauser et al (1981) [24-26].7 Laser Applications in the medical field Since their first appearance. the cavitation bubble performs several oscillations of expansion and collapses within a period of a few hundred microseconds. Photodisruption has become a well-established tool of minimally invalid surgery (MIS). Cavitation.4 the most used laser systems in medical applications are listed. Since the pressure inside the bubble again increases during collapse. It was the invention of the CO2 laser in 1967 with its high absorption by water that opened up the vast field of surgical applications. the availability of optic fibers permitted the combination of laser technology with endoscopy. and laser-induced lithotrispy of urinary calculi. The first medical application of lasers dates back to 1962. each rebound of the cavitation bubble is accompanied by another shock wave.propagates into adjacent tissue and leaves the focal volume. The time delay is caused by the material during the process of vaporization. frequently being necessary after cataract surgery. 1. some of the principal applications of laser in modern medicine are described . every collapse can induce a jet formation if the bubble is generated in the vicinity of a solid boundary. Furthermore. Approximately 30-50 ns later. In table 1. is a macroscopic effect starting rougly 50-150 ns after the laser pulse. Two of the most important applications of photodisruptive Interaction are posterior capsulotomy of the lens. 27 . Usually. when a ruby laser was used for reattachment of the retina in the eye.

These dramatic developments have reshaped the ability to treat skin disorders. café-au-lait spots. The most recent development that generated tremendous interests are the use of pulsed CO2 and Erbium lasers for skin resurfacing. i.Laser in Dermatology In dermatology. brown birthmarks and tattoos are treated by pulsed laser technology [34.e. 28 . Radiation from argon ion laser is stronghly absorbed by hemoglobin and melanin. various kinds of tissue reaction can be induced by different laser systems. Hence. five types of lasers are currently being used: argon ion lasers.. are stronghly wavelength-dependent. 28]. especially coagulation and vaporization [27. in the last years is prefered to the argon laser for treatment on port wine stains (plane angiomas). The pulsed dye laser approved by FDA in 1988 has revolutionized the treatment of this disorder and has them been used to treat other disorders of the blood vessel including broken blood vessels on the face. in particular. Erbium lasers and Nd:YAG lasers. Compared to chemical peels. absorption and scattering. Since the optical parameter of skin. Laser has recently been developed for hair removal by selectively damaging the hair bulb. Pigmented laser disorders such as aging spots. CO2 lasers. Radiation from Nd:YAG lasers is significantly less scattered and absorbed in skin than radiation from the other lasers. thermal effects of laser radiation are commonly used. One of the most frequent applications of this laser is upon patients with plane angioma pathologies The first laser to produce selective light induced injury was the pulsed dye laser. In clinical practice. these lasers resurface the skin making it much smoother and restoring it to a more youthful appearance This treatment is effective in rejuvenating skin of patients with wrinkles or other sun-induced damage and also for smoothing the textural change of acne scarring. major indications for this laser in dermatology are given by deeply located hemangiomas or semimalignant skin tumor. The optical penetration depth of Nd:YAG lasers radiation is thus much larger. This laser is thus predestined for superficial treatments of highly vascularized skin. different types of scars and even warts [29-33]. spider veins on the legs. This device is used to treat a variety of blood vessel abnormality and. By using short pulses of light that are absorbed by the upper layers of skin. The results are more impressive and side effects are no more frequent. 35]. dye lasers. laser resurfacing is more precise.

studies have established the disturbing link between periodontal disease. heart disease and strokes. Only laser systems capable of providing ultrashort pulses might be an alternative to mechanical drills. pharynx (throat) and oral cavity. Currently otolaryngologists are at the forefront of using various imaging modalities to view laser tissue interaction deep within the head and neck to afford greater therapy precision in the care of patients. 37]. Pre-surgical laser protocol can help eliminate bacteremia by efficiently addressing buildup of bacterial colonies that cause and sustain periodontal disease in a patient ready to receive open-heart surgery. The CO2 and Argon laser have been used to treat otologic disorders such as otosclerosis. ND:YAG and KTP lasers can be transmitted through a fiber optic cable to remove growths in the tracheobronchial tree as well as esophagus. many clinical studies and exstensive engineering effort still remain to be done in order to achieve satisfactory results. homeostasis and lack of swelling are necessary. Recently.Laser in Dentistry Altough dentistry was the second medical discipline (after ophthalmology) where lasers were applied. Especially in caries therapy (the most frequent dental surgery) conventional mechanical drills are still superior compared to most types of lasers. Those patients with infected gums and gum diseases suffer twice the heart attack rate and almost triple the stroke rate of those who do suffer from gum infection.Laser in Otolaryngology Otolaryngology was one of the first areas in which lasers have been applied. However. The last one is crucial when arterial cardiovascular damage is present in a patient. This allows removal of diseased tissue without compromising the patient’s breathing.. Carbon dioxide lasers have been used to remove both benign and malignant conditions of the larynx (voice box). it basically remains a field of research. These two wavelengths have been applied to surgery within the nasal passages and sinuses. Similarly. particularly CW or long-pulse lasers [36. . Particularly. The precision afforded allows delicate removal of bone tissue and replacement with artificial prothesis to improve hearing. the CO2 laser has found use in the treatment of airway in which precision. 29 .

urothelial tumors. synovitis. porokeratosis. periungual fibromas. various types of cysts including foreign body cysts. 2) Refractive surgeons use excimer lasers to reshape the cornea and help the eye focus properly. matrixectomies. farsightedness and astigmatism 3) Retina specialists use lasers (argon or ruby lasers) to treat macular degeneration. . The more common diseases include: urinary stones. hemangiomas. obstructions due to narrowing of ureteropelvic junction. plantar warts. ureteral structure. tumors. incision and excision of neuromas and release of sublesional hematomas. pigmented villondular. The procedure combines a light activated drug with a low-power red light to destroy abnormal blood vessels.Laser in Urology Various types of lasers are utilized to successfully treat a number of diseases on the urinary and genital systems [38]. Laser vision correction to treat near-sightedness.Laser in Ophthalmology In ophthalmology. 30 . various types of lasers are being applied today for either diagnostic or therapeutic purpose [39. remodeling hypertrophic scars/keloids. ingrown nails. Majority of such treatments require general anesthesia but can be performed as an outpatient basis with little or no blood loss and excellent results. However. a group of diseases caused by increased pressure in the eye. benign condylomas and superficial cancer of the penis. benign enlargement of the prostate urethral structures. . deformed nails..Laser in Podiatry Many conditions treated by laser include: fungal nails. neurofibromatosis. 40]. granulomas. skin fissures. lasers have three main applications in ophthalmology: 1) They are used to treat glaucoma. adjunct ulcer therapy (like debridement).

31 . Fiber-optically delivered lasers are being more frequently used for the vaporization and shrinkage of spinal discs to treat patients with herniated discs within the upper. mid and lowerback through minimally invasive surgical approach using the endoscopic guidance. The CO2 . particularly. if located intracranially or deep within the brain. The use of laser has been particularly beneficial for very narrow approaches that are required to access tumors in these locations.. for making incisions into the brain and spinal cord producing discrete results for the relief of pain. KTP and ND:YAG are used frequently in the vaporization and coagulation of tumors.Lasers in Neuro-surgery Microsurgical lasers are used for precision cutting.

For example. This can be done by heating the wall uniformly 32 . In active thermography a sequence of infrared images is detected during or after application of an external thermal stimulation.Monitoring of the sample under external heating In general infrared technique can be deployed in two different experimental approches depending on the specific application: passive and active method.Chapter 2 INFRARED IMAGING: Principles and applications 2. As it is known. Distinctions between the passive and active approaches are not as sharp as it seems. It is based on the recording of the natural thermal radiation emission of a body by employing a infrared camera. The result is the temperature distribution of the body’s surface under investigation. it is first necessary to assess the moisture problem. although maintenance investigation of buildings is generally considered as a passive thermography application. 42]. water in a wall activates undesired chemical reactions in building and thus a degradation process. it can be performed by the active approach as well. Infrared Thermography is a contactless imaging technique with distinct advantages. It is one among the numerous techniques used to “see the unseen''. In the case of monumental buildings. By analysis and images elaboration it is possible to achieve information about the condition of surface and subsurface structure of the sample such as: - Homogeneity degree Individuation and characterization of anomalies . Infrared Thermography (IT) stands as an attractive tool for non-contact inspections [41. Passive thermography is based on the detection of the infrared images under natural conditions without an external stimulation.1 Introduction Among the different non-destructive techniques that are in use nowadays. it uses the distribution (suffix –graphy) of surface temperatures (prefix thermo-) to assess the structure or behavior of what is under the surface. due to the extremely high cost of restoration. As the name implies.

thermographic inspection is performed at night or early in the morning and detection is based on the fact that moist areas retained day sunshine heat better than did sound dry areas because of the high thermal capacity of water. The most important features of infrared detectors are discussed and. focusing the attention on different active procedures. ln that case. the passive label is debatable since solar heating is a form of external thermal stimulation. What is generally recognized is that if one does not need thermal stimulation. finally. the procedure is called passive and active otherwise. too. ln this chapter both approaches are described. lt is interesting to note that moisture evaluation of buildings and roofs in particular has been reported by passive methods as well. 33 . In this case.from one side while observing the isotherm pattern on the other side. the chapter is completed with an overview on the new perspectives of infrared imaging applications in medical science. The thermal map recorded depends on the water content since the heat capacity is higher for water than for dry material.

abnormal temperature profiles indicate a potential problem. so that measured surface temperature (isotherm) can be related to specific behaviors or subsurface flaws. some investigations provide quantitative measurements. military surveys. where surface temperature distribution.1: Passive thermography on a sclerodermic patient: in the image hot spots associated to some morphea plaques are well visible Generally. as it is. with significant economic and quality benefits. In passive thermography.2 Passive thermography The first law of thermodynamics concerns with the principle of energy conservation and states that an important quantity of heat is released by any process consuming energy because of the law of entropy. A T value of a few degrees (> 3 °C) is generally suspicious. However. Fig. building inspection [46-49]. Temperature is thus an essential parameter to measure in order to assess proper operation. and a key term is the temperature shift with respect to a reference. 34 . and some medical investigation make use of this mode of operation. For instance. if thermal modeling is available. and so on.2. while higher values indicate strong evidence of abnormal behavior. Applications such as control of industrial processes [43-45]. passive thermography is rather qualitative since the goal is simply to individualize anomalies. the presence or absence of anomalies. etc…) [50]. often referred to as the delta-T ( T) value or the hot spot. fire detection (for example in the forest). contains relevant information concerning possible disorders. This makes it possible to optimise sewing operations through needle redesign and needle cooling. such dedicated modeling helps us to understand needle heating during high-speed sewing in the automobile industry. airbags. welding. 2. the sample is observed under natural thermal condition while no external thermal perturbation is applied. In this procedure. due to the millions of products sewed daily (seat cushions.

is based on the followed idea: If a thermal front. by images elaboration. it is necessary to generate a thermal transient inside the part. a b ACTIVE THERMOGRAPHY Fig. contrary to the passive approach. b) contrast thermal image obtained after a procedure of pulsed thermography. it is possible.2: Active thermography applied on a wood specimen: a) visible image of the specimen. such as the lack of homogeneity of the heating source. from the acquisition of a sequence of superficial images.3 Active thermography : Testing procedures ln active thermography it is necessary to transfer some energy to the specimen inspected in order to obtain significant temperature differences emphasizing the presence of subsurface anomalies.2. the active approach requires an external appropriate heat source to stimulate the materials for inspection. its part is reflected behind and modifies the temperature of the surface over the defect So. in active thermography. meet a dishomogeneous area. defect individuation and quantification purposes (inverse problem). to obtain information about structural subsurface homogeneity. travelling inside the sample. Defect detection principle. ln pulsed thermography. when step heating is used to reveal subsurface defects. during and after a thermal stimulation. This is a crucial aspect since lack of success in active termography applications is often related to problems with the thermal stimulatlon device. 2. The thermal contrast image emphasizes the presence of air holes to different depth inside the wood Hence. Such a thermal transient can be 35 .

Another benefit is that a cool thermal stimulating device will not generate any thermal noise since it is at a reduced temperature with respect to the component to be inspected. and duration. hot water jets.generated as either a static or a mobile configuration. which may be interpreted as subsurface flaws (especially in the case of the pulsed scheme.High-powered cinematographic lamps . Dishomogeneous thermal stimulation will produce spurious hot (cold) spots. Such a thermal stimulation can be either hot (hotter than the sample) or cold (cooler than the sample) since a thermal front propagates in the same way in both situations. timeliness. This uniformity characteristic is hard to achieve.Cool water jets . hot water bag The main desirable characteristics of the thermal perturbation (either cool or warm) is repeatability. uniformity.Cool water bag (with contact) More traditionally. Repeatability is needed in order to compare among results obtained on many identical parts.Cool air/gas jets . when the part to be inspected is already hot. However.Heat gun. which can be: . 36 .Sequence of quartz infrared line lamps . The cold approach has. In these cases.High-powered photographic flashes . a hot thermal perturbation is used. while cooling is more economical. Such a cool thermal perturbation device can be deployed using : . It is worthwhile to notice that lamp aging can reduce power deposit significatively: up to 40% of light intensity can be lost due to the deposition of evaporated tungsten from the filament on the inner lamp glass envelope. like the external part of a curing oven in the case of graphite-epoxy panels.Snow or ice (with contact) . hot air jets. several advantages over the warm approach: for example. under certain circumstances. Uniformity is necessary when inspection is performed on wide surfaces.laser beam (Scanned or single beam) . it is uneconomical adding extra heat to the part. described below). specific deconvolution algorithms.

Basically.1 37 . In this case. for a better agreement with theoretical models and to enable data inversion. Thus when observing the surface temperature. In this description more emphasis is given to the presentation of the pulsed termography method because part of the experimetal procedure used in this thesis work (described in the next chapters) make use of this active technique. In order to optimize this parameter a mechanical shutter may be used in front of the thermal perturbation device for the start-stop of the pulse. Consequently. deeper defects will be observed later and with reduced contrast. several active thermographic techniques exist differing from each other mainly in the way data is acquired and/or processed as reviewed and discussed briefly below. graphiteepoxy laminates) is used. In fact. Qualitatively. in fact.1 Pulsed Thermography Pulsed thermography (PT) is one of the most common thermal stimulation methods used in infrared imaging [51-62]. wood. For the pulsed scheme.specific calibration and contrast computations can allow for stimulation correction after its application. PT consists of a fast heating of the sample under investigation and then recording the temperature decay curve. Nowadays. during which a short thermal stimulation pulse lasting from a few milliseconds for high-conductivity material (such as metal) to a few seconds for low-conductivity samples (such as plastics. the phenomenon can be described in the following.3. timeliness and duration are other important issues for thermal stimulation. the observation time t is a function (in a first approximation) of the square of the depth z [63] and the loss of contrast c is inversely proportional to the cube of the depth [64] t z2  c 1 z3 2. The presence of a subsurface defect modifies the diffusion rate. 2. The temperature of the material changes rapidly after the initial thermal pulse because the thermal front propagates by diffusion under the surface and also because of radiation and convection losses. One reason for this is how fast is the inspection. a different temperature with respect to the surrounding sound area appears over a sub-surface defect once the thermal front has reached it. a square thermal pulse is needed.

transmission mode) (Fig. in a transient mode. 2.3: Pulsed Thermography set-up: reflection and trasmission mode Reflection is used when inspecting defects closer to the heated surface. depending on the thermal properties of both the material and the defect. whilst transmission is preferred for detecting defects closer to the non-heated surface (i. a pulse can be decomposed into a multitude of individual sinusoidal components. The sample is heated from one side while thermal data is collected either from the same side ( i. reflection mode) or from the opposite side (i. In that respect. It is now interesting to make the link with thermal wave theory. 2.1) show two limitations of PT : observable defects will generally be shallow and the contrast weaks. deeper defects). These relations (2. depth information is loss since thermal waves will travel the same distance whether their strength is reduced by the presence of a defect or not. A well known empirical rule says that: the radius of the smallest detectable defect should be at least one to two times larger than its depth under the surface [65]. pulse heating of a sample corresponds to the simultaneous launching into the sample of thermal waves of various amplitudes and frequencies. In general. In general.e. depth quantification is not possible in transmission. Although deeper defects can be detected in transmission. 38 .e. there are two different operation way in PT.3). reflection is most used because resolution is higher than trasmission mode and it is easier to be employed since both sides of the sample do not need to be available. Mathematically.where is the thermal diffusivity of the material. Defective zones will appear at higher or lower temperature with respect to nondefective zones on the surface. Hence. Utilization of a thermal pulse for stimulation of the sample is thus practical since thermal waves of many different frequencies are tested simultaneously (the flat frequency spectrum of a Dirac pulse).e. Fig.

frescoes. the stimulation source must be nondestructive and thus it must not damage the inspected surface either chemically or physically: this will limit its strength. Q the quantity of energy absorbed. C the specific heat. and t the time. mass density. respectively. the 39 .4 Fig. Obviously. with k being the thermal conductivity.4: Temperature evolution curve after absorption of a rectangular pulse: 1) sample made of homogeneous material. especially when high-intensity heating is directed on delicate surfaces (e.2 where T is the temperature increase of the surface. after absorption of a Dirac pulse. this temperature decay conforms to [66] T  Q e f t 2. In PT the short heating employed prevents damage to the component. In the logarithmic scale we notice that the central portion of the temperature decay follows a slope of -1/2. In fact. for a semi-infinite medium. 2. Heating is generally limited to a few degrees above the initial sample temperature. This can be an issue when high-intensity pulses are employed. and ef=(k C)1/2 is the thermal effusiviy of the material. 2.The detection phenomenon in PT is illustrated in Fig. 2) sample containing a subsurface flaw the surface temperature evolution following the application of the initial heat pulse is plotted for a thin homogeneous plate (curve 1) and for the same plate when a delamination (air layer) is present under the surface (curve 2).g. paintings of old Masters).

before and after external thermal stimulation. which are more costly. in case of the inspection of a component already at a high temperature (with respect to ambient temperature) due to the manufactured process such as for the cuing of bonded parts. can be analized pixel to pixel. 2.1 The complete thermogram sequence In pulsed termography each infrared images sequence recorded. This is attractive for thermal data visualization and quantification although it is less useful when working with phase delay images. the reading is out of the calibration scale and no accurate measure can be computed.g. Moreover. In the next section more information about a complete thermogram sequence are given. susceptible to rapid aging. the acquired thermograms are temperature saturated (2). in fact.1. In general. i. The actual number of saturated thermograms depends on the sampling frequency and on the thermal properties of the material 40 . Cold thermal sources are also advantageous since they do not generate thermal reflective noise. This approach avoids relying on high-intensity local heat sources. a cold image (1) is captured. Anyway the anomalies (such as cracks. This is.e. by means of electric heating wires). which can perturb the measurement. and potentially dangerous for the surface because of the high thermal stress they induce. The temporal evolution associated to every pixel follows a rate that can be separated in different step. the complete thermal temporal sequence for a single pixel is composed of 5 distinctive elements depicted in fig. which defines the time immediately before heat reaches the sample’s surface. it may be attractive to preheat (if necessary) the surface with a low rate of energy deposition (e. 2.3. The cold image can be used to eliminate spurious reflections due to emissivity variations and to reduce fixed pattern artifacts. this cold approach is cheaper: for instance. etc ) usually start close to the surface. ln some inspection situations. At time t0 .It is worthwhile to notice that a cold front (heat removal) will propagate in the same way as a hot front (heat deposit) inside the material.5. As for the detection depth this is limited since termography is a border technique. the basis for defect detection in active thermography as described next. During (and shortly after) the application of a heat pulse. The study of the temporal temperature profile is the core in the data-elaboration in the pulsed thermography method.

saturated thermograms. all subsequent thermograms are of interest for defect inspection and constitute the thermogram sequence (4). There are different definitions of thermal contrast in literature. 3. temperature variations are considered negligible. Deviation from the t1/2 dependency on the useful part of the thermogram provides an indication of the presence of a defective area (see fig. defects are still not visible on the ERT.under inspection: low conductivity materials saturates longer than high conductivity materials. Starting at the ERT at t1.thermogram sequence and 5. TEMPORAL EVOLUTION Fig.5 : Complete thermogram sequence of a single pixel: 1.LRT. 2. since depth is a function of time (z ~ t1/2 [42]). special care must be taken in order to perform quantitative analysis.4). 41 . 2. 4. the most used of them are described in the next section . Saturated thermograms give no valuable information and therefore can be safely discarded from the processing stage.ERT-Early Recorded Thermogram. Normally. Ideally.cold image. In order to obtain information about the presence of anomalies or defects an important parameter obtainable from an infrared sequence is the thermal contrast ( T). 2. Anyhow. will be recorded using high sampling rates. However. this condition is not always encountered in practice. The last acquired image at tN (5) corresponds to the Last Recorded Thermogram (LRT). this situation does not constitute a problem for defect detection purposes. especially when inspecting shallow defects on high conductivity materials using low sampling frequencies and/or when strong non-uniform heating is present. saturated or not. From this point. thus more thermograms.Last Recorded Thermogram The first useful thermogram that comes into sight after saturation is known as the Early Recorded Thermogram (ERT) (3).

68]. the thermal contrast is quite corrupted by non-uniform heating.2.2 Definitions of thermal contrast Several inversion procedures for defect characterization.6 . In the following these definitions are presented. Then. 70].e.Contrast for the transmission mode After the absorption of the heating energy by the front surface of the sample. 2. The difference of the two thermograms leads to the thermal contrast on the rear surface TTRANS. the DAC contrast. Each contrast definition is applied depending on the situation. . During real experiments. or for the evaluation of surface coatings) make use of normalized thermal contrast as an input signal. The temperature in the defective area shows a slower increase than in the sane area. There are some classical definitions for the thermal contrast which are acknowledged to be quite effective in reducing the effects of non uniform heating and non uniform optical properties [67. that is reference-independent and very effective in reducing non-uniform heating effects. 2.3. determination of the size. a definition of a newer contrast. is given [69. TSA(t) is the thermogram in a sane area. (i. heat diffusion within the material yields thermograms (temperature T versus time t curves) whose qualitative shapes for the rear surface are shown in Fig. at the end of the section. non-uniform emissivity and absorption distributions of the inspected sample.1. depth and thermal resistance of a defect.6 : Temperatre and contrast for the transmission mode 42 . while T(t) is the thermogram in a defective area. Fig.

heating is non uniform and specimens are subject to heat losses. which leads to: T  T TM AX  TS A TS A.3 where Sa = Sane area and MAX = maximum of thermogram.The curves in fig. eq. 2. We can see that the defective area shows a slower decrease than the sane area.Contrast-for the reflection mode After the heat excitation. .6 correspond to a theoretical case of uniform heating and adiabatic heat transfer assumption. To overcome the effect of non uniform heating. 2. which yields thermograms on both sane and defective areas showing distinct maxima. while T(t) is the thermogram in a defective area. 2. The local normalization aspect of each thermogram used in eq. In real life.7 : Temperature and contrast for the reflection mode 43 .3 is performed for each pixel of the images. 2. TSA(t) is the thermogram in a sane area. 2. The difference of the two thermograms leads to the thermal contrast on the front surface TREF Fig. M AX 2. a classical definition of the contrast [67] suggests to determine the contrast by first normalizing each thermogram on its peak value. During the processing of the infrared sequences obtained from the PT experiments. heat diffusion within the material yields thermograms whose qualitative shapes for the front surface are shown in fig.3 yields an effective attenuation of the non uniformity of the heating. then to calculate the difference between the two normalized thermograms.7.

the thermograms at those short times follow the modelling of a semi-infinite sane medium. the asymptotic level at the end of the thermograms as shown in fig. the asymptotic adiabatic temperature can be evaluated from the thermograms at a short time (i. high conductivity materials). Sa = Sane area and LRT = early recorded temperature. If we denote as TERT the temperature recorded immediately after the flash heating at time tS of the first frame.4 where LONG = normalization at long times.Similarly to the transmission case. for example.5 where tSF is the Fourier number corresponding to time tS (i. In real experiments. The curves shown in fig. For poor heat conductor materials. 44 . heating is not uniform which leads to different asymptotic levels for both defective and sane area thermograms. just after the flash heating). LRT 2.e. The evaluation of the asymptotic level T0 for the sane area is performed the same way starting from temperature T0S at time tS .7 does not exist. (2. Indeed. The locally normalized contrast for each pixel can therefore be calculated by : TSHORT  1 tSF  T T   SA  TERT TSA. TSHORT is convenient for the reflection mode when the thermal profiles do not reach stabilization (e. To reduce the effect of non uniform heating.7 are obtained under the assumption of uniform heating and negligible heat losses. low conductivity materials and/or short acquisition times).e. which leads to TLONG  T TLRT  TS A TS A. there is a definition of a locally normalized contrast. In such situations. where aZ is the thermal conductivity in the z direction).g. and then the difference between the two normalized thermograms is determined.4) cannot be used. we have : T  TERT tS F  1/ 2 2. each thermogram is normalized on its asymptotic level. 2. 2. Sa = Sane area and LRT = last recorded temperature. Such a definition is convenient for the reflection mode when a stabilization temperature is attained (e. tSF=aZ tS/e2 .6 Where SHORT = normalization at short times.ERT      2.g. and eq.

70]. which may be written as follows [66] T  Q e f t 2.8 where T(t) is the temperature evolution. respectively. therefore the local temperature for a sane area is exactly the same as for a defective area: 45 . and ef=(k C)1/2 is the thermal effusivity of the sample. Although the impact of non-uniform heating is minimized through the normalization process. The DAC technique is based on the absolute thermal contrast definition: T (t )  T (t )  TS A(t ) 2. It is also possible to use other thermal contrast definitions such as the differential absolute contrast (DAC) which has proved to effectively reduce the effect of non-uniform heating and which does not require the definition of a sane area [69.. it behaves as a sane area).7 and the 1D solution of the Fourier diffusion equation for a Dirac pulse applied on the surface (z=0) of a semi-infinite body. TSA is computed locally assuming that there is no defect visible on the first few images (i.e. At t'. The first step is to define t' as a given time value between the instant when the pulse is launched and the precise time when the first defective spot appears on the thermogram sequence. when there is enough contrast for the defect to be detected. instead of looking for a non-defective area. In the DAC method. the exact location of a non-defective area is still needed to compute the contrast. there is no indication of the existence of a defective zone yet. i. The last one is a thermophysical property relevant to transient surface heating processes measuring the material ability to exchange heat with its surroundings. Q(J/m2) is the injected energy at the surface.Differential Absolute Contrast (DAC) In the last two sections it has been presented the thermal contrast definitions for the transmission and the reflection modes (for long and short times).e.

2.10 Eq. TSA can be computed for every pixel at time t. the normalized temperature for a defect is given by : T (t ) T  T (t ) TERT 1 tS F 2. The temperature recorded immediately after the flash heating TERT at time tS is given by eq.11 and eq.9 From the above equation. lt follows that : TDAC  T (t )  t' t T (t ' ) 2. eq.12. From this definition and eq. This can be done in a similar way as described for the contrast in the reflection mode. S A  t' T (t ' ) t TERT.10 needs to be normalized. 2. S A tS F  t' T (t ' ) t t' T (t ' ) tS F tS  tS t 1 tS F 2.13 46 . we obtain a normalized definition for the DAC contrast: TDAC N  T (t )    T (t )  ERT S   tS  1 t  t  SF  2.10. 2.5. 2. 2. and substituting it into the absolute contrast definition. 2.11 and for a sound area: TS A(t ) T.12 From eq.TSA(t ' )  T (t ' )  Q e f t '  Q  T (t ' ) t ' ef 2.7.

that is. among others. Data processing includes evaluation of the thermal transit time tT.3). This technique of stepped heating (SH) is sometimes referred to as time-resolved infrared radiometry (TRIR). More recently. thanks to the availability of fast FPA-IR cameras. SH has numerous applications.3. for which the temperature decay is of interest. Variations of surface temperature with time are related to sample features as in pulsed thermography. Stepped-heating describes the functional dependence of the heating source. Experimental setup for SH expedients is similar to PT setup (fig. When SH was developed. This was because in these cases. the technique has been simplified. ceramics). techniques using pulsed heating could also be considered timeresolved. at low power).2 Step Heating (Long Pulse) In contrast to the thermal stimulation scheme.2. one of the main advantages of SH was the much faster temporal resolution achievable with respect to full-field imaging at a video rate of about 25 or 30 images/s. Time-resolved means that the temperature is monitored as it evolves during and after the heating process. The start of the heating pulse is triggered with respect to acquisition of infrared images which are recorded sequentially for later processing of the entire heating period. in case of Step Heating the increase in surface temperature is monitored during application of a stepped heating pulse (the sample is continuously heated. 2. the time the normalized temperature of a given point in the field 47 . Although strictly speaking. bond determination or evaluation of composite structures and characterization of airframe hidden corrosion. integrity of the coating-substrate. This last technique utilises thermal quadrupoles and the Laplace transform in order to derive a solution that includes the plate thickness. only infrared scanners were available.The DAC method has proven to be very efficient in reducing artifact from non-uniform heating and surface geometry [71]. for which semi-automated and fully automated algorithms [70] are available. Research is now moving towards the study of a modified DAC approach. such as for coating thickness evaluation (including multilayered coatings. At that time. An interactive step is involved in the process of determining t'. the vertical mirror of the scanner was stopped so that the device operated only in line scan mode.

Transit time tT is related to defect depth. It has been suggested [72] that for a given application. such as limitation of power for continuous sources or limitation in the adjustability of pulse duration for pulsed sources. and more specifically. have to be considered as well. the early time.2).of view differs from the evolution obtained in the case of a thick sound specimen. from a mathematical point of view. other practical criteria. 2. However. then the procedure (SH or PT) that enables to put a maximum energy Q in the specimen during tdist will produce the largest signal (eq. allows us to calibrate the image in order to take into account spatial variations of emissivity and reflectivity. 2.8: Example of curves temperature versus t1/2 for a sample with a coating of different thickness for a step heating pulse of duration 1 s. on the time scale of the phenomena of interest: slow phenomena are better explored with SH and fast phenomena with PT. fig. 2. 48 . Advantages of this approach are that defect depth and thermal properties are accessible from a single measurement without requiring knowledge about a defect-free region (such as to compute the contrast in PT). Fig. a pulse length slightly smaller than tdist can be chosen and the pulse strength can be adjusted so that the maximum surface temperature reached does not damage the specimen.8 shows the effect of the coating thickness on the sample temperature. In fact. let be called tdist the period of time required for a given defect to distinguish itself from the background. The choice to rely on PT or SP will thus depend on the applications. In general. Although from the experimental point of view SP is different than PT. both procedures provide thermal signals that contain exactly the same information. before the manifestation of subsurface defects.

2. the modulated heating). the phase image is related to the propagation time and the modulation image is related to the thermal diffusivity. For example.3 Lock-in Thermography Lock-in thermography (LT).3. enabling us to analyze the blood flow in the forearm: repetitive compression is performed through control of the air pressure going into the cuff. other heating employments are possible for various applications. interestingly. which is relatively independent of local optical and infrared surface features and hence it is interesting for nondestructive-testing purposes.g . one of the strong points of LT is the phase image. The thermographic image is a mapping of the thermal infrared power emitted. can be recorded remotely through its thermal infrared emission by an IR camera. more points allow us to reduce noise associated with the process. The depth range of images is inversely proportional to the modulation frequency of the input signal. Lockin refers to the necessity to monitor the exact time dependence between the output signal and the reference input signal (i. through sinemodulated lamp heating) while the resulting oscillating temperature field. sometimes called phototherapy radiometry. Low frequencies (0.015 Hz) reveal blood vessel visualization down to 3 mm under the skin. This is done with a locking amplifier in point-by-point laser heating or by computer in full-field employment (lamp) so that both phase and magnitude images become available. so that higher modulation frequencies restrict the analysis in a near-surface region. optical features refer to nonuniform heating. Obviously. Computer-based commercial systems of LT (such as the lock-in option of FLIR/CEDIP) works on all available points recorded over one or more modulation cycles. Applications of LT are numerous and in different fields. For example. Although four points are enough to compute A and . and conventional thermography image T.03 and 0. is performed by periodic deposition of heat on a sample's surface (e. while infrared features may concern with a variable surface emissivity. amplitude A. analysis of blood circulation was demonstrated [78].e. in the stationary regime. in addition to photothermal heating. More specifically. is based on thermal waves generated inside a specimen and detected remotely [73-77]. measurement of temperature evolution over the specimen surface permits us to reconstruct the thermal wave (travelling inside the sample) and to establish values of A (wave magnitude) and (phase) from four equidistant temperature data points. Wave generation. This has created interest in visualizing the functionality of the circulatory 49 . Such a system thus provides three images: phase . a compression cuff located in the upper arm creates the modulation. for example. In case of medicine. Here.

1889) in 1857. they generated heat and were detectable with an IR camera. In principle. making them detectable by an infrared camera. The signal generation process in vibrothermography is almost entirely determined by the interaction of the injected sonic energy with the mechanical properties of the sample. Since this effect is reciprocal. heat is released by fraction precisely at locations where defects such as cracks and delaminations are located. The thermal and IR emission characteristics of the sample play a relatively small role. a reduction in length causes an increase in temperature. the technique remained dormant until recently. E. 2. when techniques known as sonic thermography. Joule (1818. Such an idea was later formalized by the English physicists W. As a result.system and especially to discover changes related to a developing pathological process. Vibrothermography was first introduced as an application method in the late 1970’s. thanks to direct conversion from mechanical to thermal energy. In the basic scheme for modern vibrothermography-based testing. Weber (1804-1891). or thermosonic testing were introduced. who discovered in 1830 that an increase in length of a material results in a decrease in temperature. an acoustic horn is placed in contact with a solid sample. and a brief pulse of acoustic energy in the 10 – 50 kHz range is applied to the sample. Kelvin (1824-1907) in 1853 and J. Pulse energy is typically in the range of 500 – 3000 Joules. there has been considerable renewed interest in implementing vibrothermography as a nondestructive crack detection method. Periodical loading of the material thus generates hot spots at locations of stress concentration. Researchers observed that as cracks or discontinuities in a solid sample were excited with high energy – low frequency ultrasound. Significant improvements in IR camera performance over the past 20 years have made it possible to detect small cracks using lower excitation energies (or short duration pulses of high energy) than the previous reports suggested. heating of the sample is minimal unless a crack or some other source of frictional 50 . and duration less than a second. Vibrothermography (VT) is an active infrared imaging technique based on that principle: under the effect of mechanical vibrations induced externally to a structure.3. Despite this promising start.4 Vibrothermography The concept of vibrothermography (VT) originates from the German physicist W.

However. Consequently. flaws are excited at specific mechanical resonances. Nevertheless. loss of stress stop heat generation. Acoustic energy from a horn is injected into a solid sample. However. in practice. compared to other conventional active thermography methods.heating is present. Finite element modeling applying 3D equations of linear elasticity permits us to evaluate local energy concentration for components submitted to mechanical loading. In vibrothermography technique most significant advantages are the detection of flaws hardly visible by other infrared thermography schemes. while on the other side. and also at any clamping points. 2. Fig. or along the faces of a crack. causing frictional heating at the tip. local thermal gradients may appear or disappear. Local subplates formed from delaminations resonate independent of the rest of the structure at particular frequencies. This is explained by dynamic instability: one side of the hole takes more load than the other and thus becomes hotter.9 : Sketch of basic vibrothermography interaction. by changing (increasing or decreasing) the mechanical excitation frequency. there is often significant localized heating at the sonic insertion point. 51 . in some circumstances. and the ability to inspect large structural areas in situ provided that required mechanical loading can be achieved. discrimination between cracks and background features is relatively straightforward. In vibrothermography. mechanical loading provokes breakdown of the sample.

e. Even though they provide superior sensitivity and fast response time than thermal sensors. but they are less expensive and do not require cryogenic cooling as their photonic counterparts. pyroelectric detectors and liquid crystals. firstly it will be discussed the problem of the atmospheric trasmittance that define the accessible spectral bands for infrared sensor and then. thermal detectors exhibit a modest performance. the detector absorbs more energy than it radiates. an electrical current. In summary. Developments in both fields predicts that uncooled photon detectors and thermal detectors with better performance will soon emerge. Thermal detectors response is theoretically independent of wavelengths and they display high sensitivity at room temperature (see microbolometric spectral curve in fig.2. photon detectors typically require cryogenic cooling to minimize the noise to obtain the high relative sensitivity. Types of thermal detectors include thermocouples.11). The excess temperature is then removed away by conduction. with slower response than photonic sensors. some thermal and photonic detectors of interest for infrared imaging application are briefly presented. bolometers. A thermal detector responds to incident radiation by raising its temperature.4 Infrared detectors An infrared detector or sensor is a device that converts radiant energy into some other measurable form. When a thermal detector is at equilibrium there is no thermal conduction. the sensor simply radiates energy at the same rate as it absorbs it. 2. There are two main classes of infrared sensors: thermal and photonic (or quantum) detectors. 52 .g. In the next sections. pneumatic detectors. A photonic detector is made from semiconductor materials and operates on the principle that incident radiation excites electrons from the valence to the conduction atomic bands. The detector’s temperature will increase whenever the incident radiation rises above the equilibrium state.

2.4.1 Atmospheric Trasmittance

Atmosphere is a complex mix of various gases, particles, and aerosols in different concentrations. These constituents both absorb and scatter radiation as it travels from the target to the thermal imaging system. A description of the transmittance of electromagnetic energy through this mix is thus a complicated task. The main problem concerns with the transmittance of electromagnetic energy which is less than 100%. Self-absorption by gas in the atmosphere (e.g . H2O, CO2, O3) and diffusion due to particles such as molecules and aerosols are responsible for this. Moreover, various factors, such as the presence of thermal gradients and turbolences, make the index of refraction inhomogeneous and contribute further to degradation of the infrared measurement through the atmosphere. Finally, the atmosphere itself emits infrared energy which is captured by the IR sensor. Most of these effects depend on the distance, wavelength and meteorologic conditions, they are thus not easily taken into account. In order of importance, main gases that absorb the radiation transmitted are water vapor (H2O), then carbon dioxide (CO2) and ozone (O3). Due to the atmospheric special transmittance, electronic imaging system design is partionated into seven geneneric spectral regions. Four of these are associated with thermal imaging systems. The ultraviolet (UV) region ranges in wavelength from 0.2 to 0.4 μm. The visible region ranges in wavelength from 0.4 to 0.7 μm. Televisions, electronic cameras, and most solid state cameras operate in this region. The near infrared imaging spectral region (NIR) spans approximately from 0.7 to 1.1 μm. The first infrared imaging band is the short wavelength band (SWIR) which approximately covers from 1.1 to 2.5 μm. The second infrared band is the mid-wavelength infrared (MWIR), spectral region that covers approximately from 2.5 to 7.0 μm. The third infrared band is the long wavelength infrared (LWIR) spectral band. It covers the spectral region from approximately from 7 to 15 μm. The fourth infrared band is the far infrared or very long wave infrared (VLWIR) region. The last one applies to all systems whose spectral response extends over 15 μm. The MWIR and LWIR regions are sometimes called the first and second thermal imaging bands respectively. Fig. 2.10 shows a typical atmospheric transmittance curve that accounts for absorption only. The dominant absorber in the 3 to 5 μm region is carbon dioxide (CO2) which absorbs at 4.3 μm. This absorption band is obvious after a few meters of path length. Water vapor (H2O) determines the upper and lower wavelength limits for both MWIR and LWIR spectral regions.

53

Water vapor is also the dominant absorber in the LWIR region. As the water vapor increases, the LWIR transmittance decreases more quickly than the MWIR transmittance. This implies that a MWIR system may be better in a tropical or maritime environment.

SWIR

MWIR

LWIR

Fig. 2.10 : Atmospheric trasmittance and infrared imaging bands

The selection of the operating wavelength band is affected by the application. For the majority of nondestructive-testing applications, the useful portion of the infrared spectrum lies in the range from 0.8 to 15 μm, beyond 15 μm applications are more exotic, such as highperformance Fourier trasform spectrometers, which operate around 25 μm. The choice of an operating wavelength band is also important because dictates the selection of the detector type, as fig. 2.11 shows.

Fig. 2.11: Spectral response curves for some infrared detectors in different infrared regions

54

Among the important criteria for band selection there are operating distance, indoor-outdoor operation, temperature, and emissivity of the bodies of interest. As Planck's law says, high-temperature bodies emit more in the short wavelengths, consequently, long wavelengths will be of more interest to observe near room-temperature objects. Emitted radiation from ordinary objects at ambient temperature (300 K) peaks in this long-wavelength range. Long wavelengths are also preferred for outdoor operation where signal are less affected by radiation from the sun. For operation distances restricted to a few meters in the absence of fog or water droplets, atmospheric absorption has little effect. Spectral emissivity is also of great importance since it affects the emitted radiation. Although no specific rule can be formulated, the most useful bands are from 3 to 5 μm (MWIR) and from 8 to 12 μm (LWIR) since they match the atmospheric transmission band. Most of the infrared commercial products and thermographic applications fall in these categories. The SWIR band, instead, is principally used in applications where reflected infrared property of samples are important. However, detailed studies [79-81] have concluded that for temperature in the range from -10 to +130 °C, measurements can be done without much difference in both bands (from 3 to 5 and from 8 to 12 μm). For some special applications (e.g. for the military), bispectral cameras operating simultaneously in both bands have been developed to characterize target thermal more accurately.

2.4.2 Bolometers
Bolometers are thermal detectors. For bolometers, such as thermistors, the relevant physical property is the electrical conductivity. Consequently, a current must cross the element for signal generation. It should, however, be noted that this current, having its own fluctuation, adds to the detector noise. One of the main drawbacks of bolometers is in their slow response time (from 1 to 100 ms), which restricts their use to slowly varying processes. The figure of merit of bolometers is the resistive temperature coefficient defined as



1 dRB RB dT
2.14

55

The value of this coefficient is either positive or negative. Recently arrays of microbolometers have opened a whole range of new infrared imagers. defined as the slope of the material polarization P versus temperature T at the operating temperature p dP dT 2. 1 Temperature below which there is a spontaneous magnetization in the absence of an externally applied magnetic field 56 . 2. Pyroelectric detectors are sensitive only to temperature variations and thus need a chopper while modulates the incoming radiation in order to maintain a signal output.15 This effect can be improved with the application of an electrical field bias. depending on the conduction mechanism taking place within the material of interest. from l to 10% K-1 for semiconductors.1% K-1 for metal.4. A broad class of thermal detectors are pyroelectric detectors. thus causing a transient current available for pickup by the readout unit. This added complexity is a severe drawback.where RB is the internal resistence of the bolometers and T the temperature. The figure of merit of pyroelectric detectors is the pyroelectric coefficient p. unless the system is panned over the scene. Typical values are 0. For these detectors below the Curie temperature1 electric charges are generated by incident radiation absorption (heating): a change in the detector temperature generates a transient change in the surface charges. and higher values for superconductors.3 Pyroelectric detectors Pyroelectricity is defined as the property of certain crystals to produce a state of electric polarity by a change in temperature.

It peak response is near 5 μm. also called photovoltaic).2.g. the response time is short while the solid-state structure made these detectors compact. The LWIR version has a spectral responsivity from 8.HgCdTe Mercury cadmium telluride (HgCdTe or MCT) is a mixture.QWIP The quantum well infrared photodetectors (QWIP) is based upon GaAs growth technology. Consequently. and robust. supplying a signal without need for polarization. high-charge-capacity alkaline batteries can be used to induce minimum noise and ripple. The responsivity and noise are temperature sensitive so that QWIP devices are cooled to less than 60 K.4. of Hg1-X CdX Te. reliable. the spectral response can be tailored to the MWIR or LWIR region.3 to 10 μm. . For photoconductive detectors an external current is necessary to measure conductivity change while photoelectric (photovoltaic) detectors act as a power generator. A filter is used to limit the spectral response to the MWIR region. since photovoltaic detectors supply a signal by themselves. In this respect. Sinœ no heating phenomenon is needed as in case of a thermal detector. . they are much more attractive than photoconductor detectors.11 presents the spectral response curves for the most common detectors. Since biasing current is needed for photoconductor detectors.4 Quantum Detectors Quantum detectors are solid-state photonic detectors in which photon interactions either change conductivity (photoconductor) or generate voltage (photoelectric. This is essential to achieve stable results. Fig. By varying the ratio. GaAs. requiring less complex readout circuits. they are quite popular. It has generally replaced PtSi in those systems using modest sized arrays (e. InSb. A filter is used to limit the response to the LWIR region. standard video format with approximately 640x480 elements). in different ratio. 57 . The most popular is the LWIR with a peak response near 12 μm. . The wells are created by layers of GaAs/AlGaAs and the response can be tailored from 3 to 19 μm. Common materials used in photoelectric (photodiodes and phototransistors) are InAs. 2. HgCdTe detectors are used in all common module systems. PtSi and HgCdT.InSb Indium antimonide (Insb) is a high quantum efficiency MWIR detector.

a cooling unit. having the detector material and signal pathways on the same level significantly reduce the fill factor (~55% for a monolithic FPA vs. An FPA can use one of two technologies. it provides the IR-sensitive material to total surface (including the signal transmission paths) ratio. Using this technology. possibility to read out a portion of the array rather than the entire array (windowing). these devices have become popular only in the last fifteen years. FPN can be removed by recording the pattern and post processing with software to remove it [rif].) and 58 . A FPA is made up of rows and columns of individual infrared detectors. On the other hand. the amplifiers also take up more area) of the CCD is much greater. If FPAs have both IR-sensitive material and signal transmission paths on the same layer they are called monolithic FPAs.2. ~90% for a hybrid FPA). The higher the fill factor. the associated electronics. but they generally have lower performance since. all what is needed to build an infrared camera is the optics. Due yo their initial expensive cost. CMOS is expected to dominate in the low end consumer market (camcorders. the sensitivity and the filling factor (< 30%. Furthermore. possibility of developing a single-chip camera. monolithic FPAs are easier and less expensive to manufacture than hybrid FPAs.5 Infrared Imaging Devices: Focal Plane Arrays (FPA) In 1980s a new kind of imaging device started to appear and revolutionize the infrared community: large-dimensional infrared arrays simplified infrared camera construction. The readout device is also important. etc. an hybrid array has the IR-sensitive detector material on one layer and the signal-transmission and processing circuitry on another layer bonded together by small collecting bumps. The fill factor is an important parameter in selecting FPAs. snapshot cameras. since CMOS have more on-chip circuitry and uses amplifiers for each pixel. The principal advantages of CMOS with respect to CCDs are: lower fabrication cost. but this further reduces the useful dynamic range. the focal plane array (FPA). the higher the sensitivity. Similar to conventional video CCDs. toys. these chips do not require any electromechanical scanning mechanism (no moving parts) for image forming and are less fragile than pyroelectric tubes. and for some detector technologies. Focal Plane Arrays (FPAs) are today’s most common IR imaging devices configuration. producing nonuniformities known as Fixed Pattern Noise (FPN). a Charge Coupled Device (CCD) or a Complementary Metal Oxide Semiconductor (CMOS). Video signals are obtained directly by on-chip electronics drive. the cooling efficiency and the overall image quality of the system. CMOS have much higher noise than CCDs. lower voltages and have lower power requirements (portability). Hence. Conversely.

ferroelectric microbolometer arrays have performances such as time costant t on the order of a few milliseconds. detectivity on the order of 107 to 109 cm Hz1/2 W-1. . These arrays are. CCDs will remain unchallenged for scientific and technical applications requiring high fidelity. with the broader availability of large integrated-circuit fabrication technology.12 : Microbolometers array 59 . which are. and dynamic range in the next future. NETD (Noise Equivalent Temperature Difference) 50 mK. resolution. Nowaday. 2. in fact. 2. thermal masses (also called microbridges) hung by low-thermal-conductivity arms supported on metal legs anchored in the silicon substrate. modifying its electrical conductivity so that a voltage or current variation appears at the device outputs. it was possible to build large uncooled bidimensional arrays based on microbolometers. and a pixel size of 25μm (since 1024x768 array size) Fig.12). With micromilling technology.other specialty applications such as surveillance.Microbolometer Arrays ln the early 1980s. One common technology is the VOX (vanadium oxide). thus enabling a direct interface with readout electronics located under the array (fig. The IR radiation heats the hanging mass. it is now possible to constitute arrays with thousands of tiny bolometers called microbolometers. However. built using planar MOS or thin coating technologies adapted to micromilling. for example.

.1 Detectors: Performance parameters In order to compare different detectors available on the market. .16 This parameter is an important design property since the maximum power we can extract from such a detector is reached when Z  ZA 2. it also depends on the radiation wavelenght and excitation frequency. 60 .5.18 Responsivity is generally not uniform on the detector area. A brief description of some of them is given in this section.2.Impedence The detector impedence Z is an intrinsic characteristic of detectors measured using Ohm’s law Z dV dI 2. it is very useful to know the most used figure of merit for infrared devices.Responsivity Another important parameter is the responsivity expressed in voltage (RV) or current (RI) and which is the trasformation ratio of the incident optical flow F RV  dV dF RI  dI dF 2. A correct impedance match is important to maximize the output signal since electric signal at the detector output are generally small. moreover.17 Where ZA is the input impedance of the preamplifier to which the detector is connected.

to obtain a reasonable signal-to-noise ratio. receiver system bandwidth. The detectivity depends on many parameters : spectral content and modulation of the incident radiation. causing a change in the signal-to-noise-ratio of unity in the output of the detector.. NETD depends on the F-number3 of the optics and on the pixel size.Detectivity The detectivity D is also a widely used parameter to specify detectors. This the amount of power induced by an optical signal whose amplitude is equivalent to intrinsic noise power present at the detector output without signal. . . It is given by  1 2f c 2. and sensitive surface area.19 fc is the critical frequency defined by an observed decrease of 3 dB in the maximum amplitude. D should be high. . 2 3 Perfect emitter of infrared radiation with max value of emissivity The F-number of a lens is given by its focal distance divided by its effective diameter aperture 61 . detector temperature. NETD correspond to the change in temperature of a large blackbody2 in the observed scene. It is given by D 1 FNEP 2. which specifies the response time of a detector.Noise Equivalent Power A very important figure of merit for detectors is the noise equivalent power (NEP). which originates from thermal agitation and optical radiation granularity.Time costant A useful property is also the time costant .Noise Equivalent Temperature Difference Many detector manufacturers rate their detectors using the noise equivalent temperature difference (NETD) instead of NEP.20 When FNEP is the optical flow called the noise equivalent flow.

The skin temperature distribution of the human body depends on the complex relationships defining the heat exchange processes between skin tissues. The approach generally followed consisted of the detection of significant differences between the skin thermal distributions of the two hemisomus or in the evaluation of the mismatches of the skin thermal distribution with respect to an average extrapolated from a control group. after the contradictory results obtained during the 1970s. The functional infrared imaging is the study of the functional properties and alterations of the human thermoregulatory system based on the modelling of the bio-heat exchange processes.e. the action of vasodilator or vasoconstrictor substances or the execution of thermal stresses. it is possible to record in real time all the modifications of the skin thermal distribution following a physiological or environmental processes. This permits to overcome the limitation of conventional infrared imaging and to use a new approch based on the modellization of functional parameters [83]. The presence of a disease can locally affect the heat balance or exchange processes resulting in an increase or in a decrease of 62 . During the last 30 years it has been widely discussed whether or not infrared imaging should be considered a good diagnostic tool. but for other purposes. a general lack of confidence has arisen among clinicians regarding the clinical use of infrared imaging. technological improvements released a new generation of digital infrared cameras characterized by : - high spatial resolution optimized temperature sensitivity very short acquisition time By means of the last generation infrared cameras. So. and metabolic activity. inner tissues. All these processes are mediated and regulated by the sympathetic and parasympathetic activity to maintain the thermal homeostasis. Simultaneously.2.6 New approach of infrared imaging in medical science Infrared imaging allows the representation of the surface thermal distribution of the human body. A dynamic approach to infrared imaging by means of the characterization of the dynamics of the thermal recovery is therefore possible [82]. Several studies have been performed so far to assess the contribution that this kind of information may give to clinicians. i. local vasculature.

The submotor thermal response may he used in the future as complementary or alternative monitoring of the activity of the autonomous system in conditions where the traditional sympathetic skin response cannot be recorded. Such a temperature change can be better estimated with respect to the surrounding regions or the unaffected contra lateral region. to determine the major infrared thermographic parameters in discriminating between patients with and without secondary Raynaud’s phenomenon. the characteristic parameters modelling the activity of the skin thermoregulatory system can be used as diagnostic parameters. the dynamics of the local control of the skin temperature should be also influenced by the presence of the disease. 87] - Validation and detection of sub-clinical varicocele [88-90] Moreover. The method is based on a bioheat transfer model that reflects the thermo-physiological processes in a skin region proximal to a major vessel [96]. Therefore. Infrared technology is used for: - Valuation of re-perfusion process during surgery (thermo-angiography) [84] Follow up of the modification of the neurovascular functions of the treated district by means of the study of thermoregulatory properties [85. by evaluating qualitatevely and quantitatively how the lesion or the disease alters the thermoregulatory response of the damaged or ill area with respect to the healthy ones. 86] - Assistance to cardiac surgery and microsurgery. also. It is based on the 63 . Infrared Imaging was used to record the temperature variation of the hands associated with sympathetic skin response [91]. The aims of this new approach of infrared imaging is highlighting the functional contents of skin thermal distribution. for wich the real time evaluation of the re-perfusion processes and the estimate of the ricovery of the neurovascular functions constitute important information during the follow up [84.the skin temperature. Some of these regard the Raynaud’s phenomenon secondary to Scleroderma (SSc) or chronic vasospastic syndromes [92-95]. But then. a novel method of estimation of blood flow speed and vessel location from thermal imaging recording has been developed. Recently. Several studies have been performed so far to assess the contribution that this new approach may provide to the medical science. The new infrared approach is used. A new possibility on the remote control of the humans functional parameters is given from a novel method to measure human cardiac pulse at a certain distance.

All these studies (and many others) has proved that infrared thecnology can be advantageously used in these field thanks to its new features. sport training. Some studies show how as the infrared imaging. The contribution that the functional infrared imaging can give in the diagnosis of the breast cancer has not been clarified.information contained in the thermal signal emitted from major superficial vessels [97-99]. overall thanks to new dinamyc approach and new images elaboration methods. sleep studies. The technology is expected to find applications among others in sustained physiological monitoring of cardiopulmonary diseases. thanks to pattern recognition methods. 64 . Anyway. and psychophysiology (polygraph). the capacity to localize the tumor with this technique is subordinates to the realization of new realistic model of the thermal diffusion in vivo [100-102]. can individualize the tumoral neoangiogenesis and the excessive production from endotelium of nitric oxide associated to the lesion.

The scientific research in this field is addressed to obtain quantitative and functionals informations about the dynamic of this interaction in order to allow to specialists to employ laser systems in a more specific. the approach employed.biological tissue interaction on patients affected by plane angioma pathology has been monitored and studied with the infrared imaging technology. Subsequently. they are unsatisfactories or even harmfuls. 65 . This activity has been realized in collaboration with the LIRT of the Cybernetic Institute “E.1 Introduction In the first chapters the problems concerning to laser applications in medical fields and infrared imaging technologies have been described. object of the present thesis-work. the employment of differents laser systems for more and more medical treatments has had an esponential growth. instead. in the majority of cases.ICIB-CNR and in collaboration with the dermatological division of the II Policlinic in Naples in the laser therapy and photodynamics ambulatory. 3. the pathology analized and computation instruments used (pulsed thermography model and a numerical simulation) will be treated and discussed. and. in which the laser . it will be given a detailed description of the activity research. The effects of the laser radiation on differents organics tissues aren’t easily standardizable.Chapter 3 INFRARED MONITORING ON PLANE ANGIOMA : PT model and numerical simulation 3. This indetermination influences the therapeutics results obtained. well controlled and less empiric way. In this chapter. because of dermatological differences showed from differents anatomics regions of the body and in differents people.2 Infrared monitoring of plane angioma treatment: approach description In the last years. Caianiello” .

This technique can give to specialist information about the reaction of the tissue in the specific region treated. In this activity the main pathology monitored and studied with infrared technique has been the Plane Angioma known as Port Wine Stain (PWS) pathology. Usually. - Possible formation of scars - Possible formation of dischromic area In the laser treatments on patients affected by Plane Angioma the therapeutic program and laser parameters are chosen following general tables and indications present in literature. phototype. in this field. as described in the next section.7 mm Tab. lesion depth and lesion thickness. is a vascular malformation that consists in a blood vessels accumulation under the tissue. anatomic region. in general. This accumulation forms a subsurface plane of vessels of depth and thickness depending on the specific anatomic region and. in the median region of the face the therapeutic efficiency is better then external area.Infrared imaging technology can be used. The main problems in the cure of plane angioma with laser therapy are: - The therapeutic efficiency can differ a lot from patient to patient. In the next table it is shown the reference range for laser parameters advised in literature for the treatment of patients with Plane Angioma: Referential range for laser parameters Wavelength 585 – 595 nm 4 – 10 J/cm2 1 – 5 ms Energy density Pulse duration Spot diameter 2 . for a real-time monitoring of the laser-biological tissue interaction. 3. differents from case to case. In general it can depend on the age. This pathology.1: Laser parameters for Plane Angioma treatments 66 .

object of this work-thesis. it is possible to get a local evaluation of the depth and thickness of the plane angioma in the anatomic area that the physician wants to treat. The main purposes of this study have been: - To evaluate the support of the infrared imaging. With this technique it’s possible to obtain information about a specific plane of vessels under treatment monitoring the thermal reaction of the tissue under laser action. 67 . In the simulations the thermal behaviour of the biological multilayers can be achieved with different set of laser parameters as input. In general. with the intent of optimazed laser parameters for each specific plane of vessels treated by specialist. from the temporal evolution of the surface’s temperature obtained after a single laser pulse-test (heat souce). in order to find morphological information for the specific plane of vessels under treatment a Pulsed Thermography model has been used. the next approach has been studied. the effects of the laser-biological tissue interaction on PWS patients have been studied through the information obtained with infrared imaging technique. In the research activity. Subsequently. in fact. This information permits a morphological reconstruction of the multilayer biological tissue under treatment. in the dermtological field. the specialist starts from the laser parameters showed in the table and then he changes these parameters based on his own experiences and so in a very empiric way. First of all. With this model. nowadays it is the operator who chooses the system and the operative protocol. as an efficient tool for scientific investigation And particularly - To evaluate the employment of this technique as a tool for the control and optimization of laser parameters in order to make the treatment more safety and more efficient So.Even if modern laser system permitted more and more control on the operative variables. it is executed on this multilayer a 2D+1 numerical simulation based on a biometric heat transfer model.

Variation of the laser parameters in order to optimize the local temperatures obtained .Recording of the temporal evolution with infrared camera .Numerical simulation on the biological multilayer starting with the referential set of laser parameters .The numerical output consists of the evaluations of the spatials and temporal temperature profiles.To compare the thermal behaviour simulated on the multilayer’s surface with the experimental data (recorded with a infrared camera) in order to check the approach performed.Reconstruction of the biological multilayer under treatment . Finally. from the temporal evolution recorded .Comparision of simulated and experimental results { 68 .To optimize the laser parameters in order to achieve suitable local temperatures in the multilayer for the photocoagulation of the blood vessels and for reducing the risk of scars formation Schematically. this approach can be synthesized in the next operative flow-chart : { { { .Evaluation of the plane angioma depth (d) and thickness (h) by PT model. . these results can be used: .Local temperatures evaluation .Single Pulse test on the plane angioma region under treatment with the reference set of laser parameters .

the pulsed thermography model and the numerical simulation used in this approach are described and discussed extensively.As it is possible to see from the flow-chart. In the next sections the plane angioma pathology. the results achieved from simulations are used as a feedback in a loop in order to optimize laser parameters and so to optimize the therapeutic efficency. 69 . in this operative approach.

In adulthood. PWSs are present at birth and persist throughout life. thickening of the lesion or the development of small lumps 70 . Early stains are usually flat and pink in appearance.3 Plane angioma pathology (Port Wine Stain) In the last 20th years. 3. acquired PWS can occur at any age after birth and it is identical to congenital capillary malformations both clinically and histologically. Plane angioma occur most often on the face but can appear anywhere on the body. Evidence for genetic influence is lacking. The mean dimensions of the blood vessels in a plane angiona are reported in the next table Blood vessels dimensions Thickness 20 – 180 μm 200 – 1300 μm Depth Tab. Rarely. This vascular malformation. is congenital and it is present in about 3 newborns on 1000 with the same incidence for male and female. The exact mechanism (causes) remains unknown. the plane angioma is a benign tumor made of blood vessels accumulation newformed and malformed.3. As it is well known. the color may deepen to a dark red or purplish colour. Capillary malformations may result from a neural deficiency of sympathetic innervation of the superficial dermal blood vessels.2 : Mean dimensions of the blood vessels in a plane angioma The area of the skin affected grows in size commensurate with the general growth. The color of the capillary malformation does not correlate with the capillary depth or diameter. one of the field in which is much more concentrated the dermatological interest has been the treatment of the skin vascular lesion as plane angiomas (knowen as Port Wine Stain or Nevus Flammeus too). As the child matures.

A study demonstrated that the psychosocial difficulties not only persisted but actually worsened in adulthood [103]. in approximately 20-25 % of cases there is an optimal improvement. Older stains may be more difficult to treat. and limitations of privileges and opportunities otherwise afforded to those without facial disfigurement. In unusual cases. isolated and passive orientation in interpersonal relationships. radiation. Parkes-Weber syndrome. loss of function (especially near the eye or mouth). and tattooing. Wyburn-Mason sindrome and others. (plane angioma can also be covered with cosmetics).may occur. However. the presence of PWS cause emotional and social problems for the affected person because of their cosmetic appearance. stigmatization from society. During the years many type of laser system have been used: CO2 . and increasing disfigurement. Treatment of infants with the FPDL generally produces marked improvement in appearance. ruminative self-doubt in interpersonal interactions. Lasers have had the biggest impact on treatment. 71 . in general. Nd:Yag and others. complete disappearance is rare. because they are the unique method for destroying the cutaneous capillaries without a significant damage to the overlying skin. If the PWS is on the face or other highly visible part of the body. These complications are usually seen later in life and psychosocial disability secondary to facial disfigurement can be overwhelming. Proteus syndrome. Argon laser. Von Hippel – Lindau syndrome. hypertrophy (increased tissue mass) of the stains may occasionally produce deformity. the laser system tool more efficient and less dangerous for curing PWS patients. Several studies demonstrate that patients with facial capillary malformations exhibit greater self-concern. Klippel-Trenaunay syndrome. A physician can usually diagnose a PWS based entirely upon the history and appearance. in approximately 50% of cases the therapy is less incisive and there may be no improvement. a skin biopsy may be needed to confirm the diagnosis. Many treatments have been tried for PWS including freezing. In same cases (about 5%) plane angiomas can be a cutaneous finding of several syndromes as: Sturge-Weber syndrome. social inhibition. Cobb sindrome. bleeding. Stains on the face respond better than those on the trunk or limbs. In the absence of successful treatment. in 30-35 % the improvement is acceptable and. Usually. unfortunately. but several study and experiences have highlighted as the FPDL (Flash-Pump-Dye-Laser) is. at present. surgery.

4 Pulsed Thermography model In this work.3. is thermally characterized by a thermal contact resistance. then a double Fourier transform in the space domain is applied to the previous Laplace temperature field. in this model. 57]. from the temporal evolution of the superficial temperature. in order to find unknown parameters for each specific subsurface plane of vessels. the role of subsurface anomaly is rappresented by the plane of vassels associated with the presence of a plane angioma [rif ben]. Resuming. with this approach. it has been used a tested modelling approach of pulsed thermography (PT) based on the use of the 3D thermal quadrupole formalism [56. simply and rapidly as possible 2) The second step is to develop an inverse problem that estimates the unknown parameters in order to minimize the deviation between the experimental data and the forward problem These two important steps will be accurately described in the next subsections. In the multilayer biological structures analyzed in this work. two stages have been considered: 1) The first step is to develop a forward model which consists of describing the contrast field evolution in a mathematical formalism as accurately. A plane of blood vessels. in a multi-layer biological tissue. in order to accurately identify the unknow parameters associated to anomaly. The analytical solutions obtained via this method are very convenient for the development of new inversion procedures which characterize the subsurface anomalies within a three-dimensional heat transfer configuration. three-dimensional thermal quadrupoles are obtained by applying to the real temperature field a Laplace transform in the time domain. In this model. After the last three integral transforms. 72 . the temperature field is treated in the Laplace–Fourier–Fourier image domain. The perturbation procedure is based on an asymptotic expansion of the physical field (temperature or heat flux) versus a small parameter intervening in the model. it is possible to estimate the depth and the thickness of a subsurface anomaly or disomogeneous area present in a structure. In general. Because this resistance is small compared with the tissue whole resistance. a perturbation method [104-106] can be applied and combined with the thermal quadrupole formalism to yield simple analytical solutions of the nondestructive experiment.

24 riferim]. First of all. the quadrupole technique is the easiest formalism to implement in order to describe the heat transfer. Heat transfer modelling of multi-layer materials is then reduced to a simple multiplication of matrices in the transformed Laplace–Fourier–Fourier domain. 3.1 73 .4.1 Formulation of the forward problem: 3D thermal quadrupoles This model refer to the case of a rectangular (L1 × L2) multilayer sample of thickness d1 that contains a resistive anomaly of finite width a and finite length b. Fig. recall here the heat equation in 2+1 dimensions: kz  2T  2T  2T T  kx 2  k y 2  0 t z 2 x y 3.1: Multilayer sample of thickness d1.the front face of the sample is adiabatic 3 . one can assume that: 1 . and a double Fourier integral transform on the space variables x and y [23. with a plane angioma depth d and thick h v In a typical pulsed infrared thermography test.the thermal excitation of the sample is a Dirac distribution characterized by a uniform absorbed energy Q (at time t = 0) 2 . typical of biological tissue affected by plane angioma.the sample temperature is zero before excitation Since only the face temperatures are of interest.3. with a uniform contact resistance Rc on its whole area. Thermal quadrupoles are obtained after performing a Laplace integral transform on the time variable t.

the Laplace transform τ(x. z. z. d is the depth of the anomaly. Superscripts sup and inf relate to the upper and lower face of the anomaly z  d1   0 z 3. y. y2 )] and s(x. p) (where p is the Laplace variable) of the temperature T (x. y) inside the anomaly region [(x1. t) in the sample is the solution of the following set of equations:  2 k x  2 k y   z 2 k z x 2 k z  x  0. y) = 1 for (x. y) = 0 elsewhere.where T is the temperature and k is the thermal conductivity. If the initial temperature is uniform and if the heat losses are neglected. L2  y z0 zd  k z  2 p   0 2 kz y 0 0 3.4  1 2 * Rc  Rc d1 k z  z *  z d1 d *  d d1 74 .2  sup   inf  Q z  sup  inf   z z     Rc s x. L1  x  y  0. (y1. x2 ) .3 The analysis can simultaneously be performed for both isotropic and anisotropic materials (providing that the sample faces are parallel to the main directions of anisotropy) by using the following dimensionless governing parameters:  *   Qd1 k z   *  Q x*  y*  x k z k x 1 2 d1 y kz k y d1 a*  b*  p  * a k z k x 1 2 d1 b kz ky d1 d12 p kz  1 2 3. y  k z z    with s(x. y.

The partial differential equation in 3. the asterisk superscript in the above parameters will be omitted for clarity reasons. p  cosx  cosy dxdy 0 0 L1L2 3.where ψ is the Laplace transform of the z component of the heat flux density =-kz ( T/ z). 75 .5 The next step is to apply a Fourier transform to the function τ (x. back inversion into the Laplace domain is given by  x.6) and the lateral boundary conditions. z.5) yields j and k are d 2  p  2   2   0 2 dz   3. The lateral boundary conditions determine the type of transform to carry out [25]. y.8 where u =( p + α2 + β2)1/2 and F and G are two constants that can be determined using the boundary and interface conditions in z. non-negative integers. In the remaining text of this chapter. y. βk = kπ/L2. z. If θ is known. equation (3. p). they determine the discrete values allowed for the space frequencies: αJ = jπ/L1. in this case it is a cosine shape q( .9 where θjk = θ(αj. z. z. z.  . p        1  00  2  j 0 cos( j x)  2  0k cos(  k y)  4    jk cos  j x cos  k y  L1L2  j 1 k 1 j 1k 1    3. p).7 The solution of equation 3. Using equation (3.6 As for the boundary conditions on the opposite faces x = L1 and y =L2.7 has the following form:   F coshuz   G sinhuz  3. y. p)     x.2 then becomes  2  2  2    p  0 z 2 x 2 y 2 3. βk.

If

is the Fourier transform of the Laplace heat flux density ψ, which is unity at z = 0 and zero

at z = 1 (equations 3.2 and 3.3) we obtain

z0 z 1

 

  ,  ,0, p  

sin L1  sin L2 

3.10

  ,  ,1, p   0
, this

If the arguments other than z are omitted in the Laplace–Fourier transforms θ and

equation leads to a linear relationship between the two quantities on the front (z = 0) and rear (z = 1) faces

 0  A 1  B 1  0  C 1  D 1

3.11

Boundary conditions in (3.2) and (3.3) can be rewritten under the following matrix forms, commonly called quadrupoles [107]:

 0   A  sin L1  sin L2     1 C    1     sup  sup    Rc I      d    d    inf   A2 B2   1      d  C 2 D2   0 
x2 y 2 x1 y1

B1   sup    D1   d  
3.12

I    x, y, d , p  cos x  cos y dxdy

where
Ai  Di  cosh ud i  Bi  1 sinhud i  u Ci  u sinhud i  for i  1,2

where d1 and d2 are the thicknesses of the two layers of the sample. A simpler and faster method to obtain a solution of the equations (3.12) is the perturbation method. In the following section, this approach is presented and analyzed.

76

3.4.2 Resolution of the forward problem: mathematical perturbations
The perturbation method consists of writing asymptotic series expansions of the variables θ and with respect to a small parameter in the model. These series expansions are then injected into set (3.12), and term-by-term identification of the coefficients of the successive powers of the small parameter leads to a coupled series of linear sets. These allow the calculation of the different components θi and
i.

The mathematical formalism presented below is valid only for

small values of the thermal contact resistance Rc. The asymptotic series expansions with respect to the small parameter Rc that will be noted ε from now on are

  ,  , z, p     i  ,  , z, p  i
i 0 

  ,  , z, p    i  ,  , z, p 
i 0

i

3.13

Term-by-term identification of the coefficients of εn leads to the following quadrupole equations: For ε0-order identification set (3.11) results in

 0   A  sin L1  sin L2    1   C1     

B1   A2 D1  C 2 

B2   0 1 D2   0   

3.14

 0 e1   sinh

pe2 / sinh

 p

For ε1-order identification set (3.12) turns out to be

1 0   A1  0   C    1

B1    sup    D1  1 d  

1sup  1inf  I 0     1 d   1 d       inf   1  A2 B2  1 1     1 d  C 2 D2   0   

3.15

77

I0 has the same definition as I in equation (3.12), just replacing ψ by ψ0. The substitution of ψ0 defined by equation (3.14) into this definition allows the calculation of integral I0 and therefore the solution of set (3.15). The first-order temperature on the front face is given by:

 0 

4



K

sinh(ud 2 ) sinh( p d 2 ) sinh

 p sinhu 

3.16

where

 x  x1   x2  x1   y 2  y1   y 2  y1  K  sin  2  cos   sin   cos   2   2   2   2  

3.17

Product of equation 3.16 by ε represent the Laplace–Fourier transforms θ = εθ1 of the contrast T(x, y, zs, t) on the front face. Return to the original (x, y, zs, t) domain can be achieved numerically by using fast Fourier-transform (FFT) and Stehfest algorithms [108]. Let us remind that the modelling described previously is based on the assumption of an adiabatic heat transfer. In reality, nondestructive experiments are always affected by heat losses. However, as reported in the literature [109], the losses effect on thermal contrasts are very small. This last result is a consequence of the nature of the thermal contrast itself: even if each surface temperature T and T0 might be strongly affected by heat losses, their difference, T = T − T0, leads in general to a compensation for this effect. Numerical simulations have been carried out in a previous work [106] to assess the effectiveness of the above model. The simulations have shown that the perturbed contrast profiles are in excellent agreement with the true ones for small thermal resistance values, with ε 0.3. However, for higher resistances (ε > 0.3) the approximation is no longer appropriate,

confirming the restriction of the above model to weak anomalies or defects. We can observe, that in the case of anomaly rappresented by a plane of vessels, characteristic of a plane angioma, this condition is accomplished.

78

kt=0.087  0.492 10-8 W m-1 K-1 .3 ht k v 3. in the case of a plane of vessels thick hv=200 μm. the normalized thermal resistance ε is given by  hv k t  0. it is described the inversion procedures that allow the evaluation of the depth and the thickness of the subsurface plane of vessels.In fact. These parameters are directly derived from the perturbation 3D modelling described in this section.5 mm (maximum depth of laser interaction in a biological tissue for a wavelength =595 nm) with a thermal conductivity for the plane of vessels and for the tissue respectively of kv=0.18 In the following sections.322 10-8 W m-1 K-1. 79 . inside a tissue thick ht 1.

21 mi   x. the depth d of the flaw is the solution of the following equation: cosh where  m  p1 1  d    2  m   1  12 cosh  p 1 3. y. p    sinh 2  p 1  d  sinh 2  p  3. With the previous firstorder perturbation method.16) allows the calculation of τ for the front face:  0.3 Evaluation of the depth In this section.zs .19) is equal to εθ1(0. p).22 80 . p). z s . pi  and i  1.4.0.20) written for two values p1 and p2=4p1 yields the analytical elimination of the unknown parameters ε and.19 where τ is the difference between the Laplace transform of the temperature on one spot of the face of the sample and its value in the absence of anomalies. p)dxdy 0 0 L1L2 3. The space average of the Laplace contrast is defined as      ( x.3. y. the integral in the right-hand side of equation (3. it is shown how the depth of the anomaly can be derived from the space averaging of the Laplace contrast profile τ(x. y. If m1 and m2 are the experimental average Laplace contrasts on the front face.2 This equation leads to the following estimation of the anomaly depth: d  1 1 m  ln[ 2  p1  m1    12 cosh   m p1   2 cosh 2 m  1    p1  1 ]   12 3. The application of equation (3. 0.20 Equation (3. zs. In the case of expansive plane of vassels ab/L1L2=1.

it is shown how the thickness of the anomaly can be derived from the space averaging of the Laplace contrast profile τ(x. the parameter d can then be eliminated from the two corresponding equations by using the duplication properties of the hyperbolic functions. kv is simply the thermal conductivity of the blood. p). ε.4.4 Evaluation of the thickness In this section. zs. If it is written for two values p1 and p2 = 4p1 of the Laplace variable. if the depth of the anomaly d has been already calculated. y c . The equation (3.0.20) shows that the average Laplace contrast (and therefore the instantaneous averaged contrast τ ) is proportional to the thermal resistance ε of the flaw. the thermal resistance of the plane of vessels is given by the following equation:  2 m1 sinh 2 m2  p tanh p  cosh  p   m 1 1 1 1 3. Evaluation of the thickness hv is therefore equivalent to the estimation of its thermal resistance. pi  and i  1.3. ε =hv / kv.2 81 . If m1 and m2 are the experimental Laplace contrasts on the front face at the centre of the flaw.23 where mi   xc .20 depends on two unknowns: the thermal resistance ε. The thermal contact resistance per surface unit is defined as the ratio of the anomaly thickness hv and its thermal conductivity kv. In other words. and the depth d. which is a known parameter. y. In case of anomaly rappresented by plane of blood vassels. Equation 3. the average Laplace contrast is proportional to the thickness of the plane of vessels.

this second error will be neglected since the numbers of points. of the thermal resistance and of the depth are reported as a function of the Laplace variable p (= p1 ) for different values of the anomaly depth.25 2 1 2 1 2 In figure 3. This error can be divided into two parts: the error caused by measurement noise.25. and the error caused by performing the quadratures in time to calculate the Laplace transform and in space to calculate the average Laplace contrast. on a limited number of points.4.25 are 1   2 p1  2   2 2 t UV   1 p p 2  1 1 p1  p 2 1 2 p1   U 1     V1      U1  V1  p   m  m cosh  p  1  m p sinh p cosh  p   1  m p sinh p   m  m cosh  p  m1 1  m1 p1 sinh 1 1 1 1 2 1 1 1 2 1 2 1 2 1   m1  2m2 cosh  p 1 3. The effects of measurement noise on the precision of the thermal resistance and depth represented by the variance equations 3. 82 .24 2 d  2 t 1    cosh 4 NM 2 p1     p 1  d  1   exp    cosh p1 1  d   cosh   p 1  d    p 1  d   2    exp  p 1  d    1       2 U 2    V   2 U2  V2 p 2 p sinh  p 1  d  2 p sinh  p 1  d  2 sinh  p sinh  p    p  p sinh  p 1  d sinh  p 1  d  sinh 4 4 1  p 1 2  sinh 4 4 1 2 2 2 2 1 2 3. the relative standard deviation. σd/d.2. A measurement of the quality of the inversion requires a quantification of the estimation error that has been made. Using a stochastic analysis with the assumptions that the noise on the thermal contrast T is additive.5 Statistical analysis of the accuracy of the inversion procedures Any inversion algorithm produces estimated parameters. we used a standard deviation of the normalized T contrast σ = 0.24 and 3. it is possible to quantify the inversion errors caused by measurement noise.2 the value d1 refers to maximum length of laser interaction in the tissue ( 1.24 and 3.3. In the following. σ /ε.5 mm for a laser wavelength =595 nm). In the figure 3. uncorrelated and of constant standard deviation σ. used to perform the quadratures are considered large enough. With regard to the other parameters in equations 3. in time and space.1.

1d1 up to 20% for d = 0. a) b) Fig. but it is always in the range 1–2 for deep anomalies.1d1 up to 75% for d = 0.7d1.b 1) The optimum p-value that should be used for thermal resistance inversion independently from the anomaly depth must be within the range 1–2. 2) A bad choice of p leads only to a small error on the precision for plane of vessels located near the heated surface. a) σd/d for the depth. From fig. 83 . 3. whereas for deeper plane of vessels.087.2. 3. 3. 4) the estimation error changes nonlinearly with the Laplace variable p and the flaw depth d.a 1) The choice of the optimum p-value depends on the depth d.2: relative standard deviation as a function of the Laplace variable p (= p1 ) for different values of the anomaly depth.a normalized time step t = 4 × 10-3 that corresponds a scan rate of 40 ms. a normalized thermal resistance ε = 0. the numbers M and N are both 256. 2) The estimation of the thermal resistance is better for more superficial plane of vessels.7d1. for optimum p-values the error changes from 25% for d = 0. the discrepancies are higher. b) σ /ε for the thermal resistance We can note from the figures that: From fig. 3) For the deep anomaly the error can changes from 3% for d = 0.2.

This model describes heat transfer in the vicinity of a large plane of vessels proximal to the skin. The plane of vessels can be considered a single large vessel running along the x direction without traversing across the z direction (see Figure 3. in positive z direction. Fig.3.3). We can neglect heat transfer along the (y) axis because of the presence of other vessels. Therefore. and specific heat c of the tissue. z   q L x. These layers are successively.3: Multilayer biological tissue It has been also assumed that each layer is isotropic with respect to thermal conductivity k(z). and the dermis again (core). obtained for different plane of vessels with PT approach. have been used as input data for a numerical simulation based on a biometric heat transfer model. metabolic heat rate qM. z  t x  x  z  z  3. 3. The mathematical problem is to retrieve the unknown spatial and temporal temperature profiles associated with a specific multi-layer biological structure and for a given set of laser parameters. the 2D + 1 model assumes the following form: c T   T    T    k z     k z    q BL x.26 84 . The heat effect of the vessels on the skin temperature depends on the vessel’s location (depth) and from its thickness as well as the blood flow speed and temperature. periodically arranged and similar to that considered. The thermal conduction in the tissue surrounding the plane of vessel is dominant in directions parallel (x) and perpendicular (z) to the skin. the plane of vessels. the epidermis. density ρ. It is assumed that the vessels act as a volumetric heat source for the surrounding four-layer tissue structure. the dermis.5 Numerical simulation: biometric heat transfer model The morphological informations. t   q M x.

The heat source term associated with blood flow is assumed to have the decomposition qBL = uBL(t)r(x. In the model.28    BL c BL A Tvessels x. z . qM is the volumetric metabolic heat.4 (vair)0. D.0. z) the modified bell function:  z  d 2   r x.0. z).0. µ is defined as follows: 3. z    exp    hv 2    hv is the thickness of the plane of vessels seen as a heat source. z. where vair is the air speed in (m/s). L  x  0. t  T ( x. k(z) is the thermal conductivity of a particular layer. where uBL is the unknown blood flow speed in the plane of vessels and r(x. where σ is the Stefan-Boltzmann constant and is the skin emissivity. t   Tair   qir z T (0 / L. t )  Tepidermis x. z. qBL is the heat due to mean blood flow speed uBL in the plane of vessels.27 T ( x. qir is the radiation heat flux: qir = σ (Tepi 4 − Tair4). t )   T x. A is the vessels cross section. t )  0 z  0.67 (W/m2 K). and V is the control volume of tissue. the following boundary conditions are used: T ( x. t  V 3.7 + 7. We assume that the early temperature of the blood in the vessels is the same as the core (dermis) temperature Tvessels =Tdermis . The heat source term associated with laser source is assumed to have a gaussian expression 85 .29 where ρBL and cBL are the density and the specific heat of blood respectively. D  x λ is the convection heat transfer coefficient. L  x  0. which depends on air flow. t )  Tcore x  0. while ρ and c are the tissue density and specific heat respectively. L  3.where qL is the heat from the laser source. According to [110]: λ = 2. t   T x.

The main advantage of the finite volume method is that it is easily formulated to allow for unstructured and dishomogeneus meshes (structures).31          86 . using the divergence theorem. This problem is well posed and has a unique continuous solution. these methods are conservative. In constructing this model. j 1 / 2  in. is the laser wavelenght and tp is the pulse period. it has been assumed that there isn’t heat flux between the domain of interest (0. In this method the values are calculated at discrete places on a meshed geometry. j  i. a is the total attenuation s is the absorption coefficient. The solution is C1 but has second order derivative in the z direction with finite jump at the line of discontinuities z = Ct of the thermal conductivity k. "Finite volume" refers to the small volume surrounding each node point on a mesh. j  in11/ 2. . The finite volume method is one of the methods used for representing and evaluating partial differential equations in the form of algebraic equations.j the average value T of in the centered FV cells.26) is: Tinj1xz  Tinj xz  .1 1 / 2  j . In the finite volume method.5.  D   BL cBL  BL cBL  BL cBL VASO   3. L) × (0.  z h2 2 tz tx n 1 txz  ( uBL ) 1  Tinj1 exp  j 2   q0e ( z) z ex w( z)  q M in11/ 2. volume integrals in a partial differential equation that contain a divergence term are converted to surface integrals.D) with the rest of the body. Let us denote Ti.30 qL  0   a   s 1  g   2   2  w( z ) 2  w0 2 1   z w0         Where q0 is the central maximum depending on the laser fluence. is the scattering coefficient. g is the anisotropy coefficient. The discrete version of equation (3. These terms are then evaluated as fluxes at the surfaces of each finite volume. w(z) is the waist size. Because the flux entering a given volume is identical to that leaving the adjacent volume. 3. w0 is the half-spot diameter.1 Model Computation In order to obtain a direct simulation of the model it has been used a Finite Volume (FV) approximation with centered cells of size hx × hz on a regular space grid. with q L  q0e ( z ) z e x w( z )  2 0  t  tP t  tP 3. coefficient depending on the biological layer.

The numerical computations have been realized using the Matlab 6. j hx Ti . j=1.Nz column-wise or row-wise. j 1  k i .5 code. xi  hx / 2  z j  hz / 2  i 1 hx N x 1 j 1 hz N z 1 i  1.35 At each time step we have to solve a linear system to obtain Tn+1(x. j  k i 1 / 2. j 1 / 2 1 k i 1. The grid computation is fixed to Nx=Nz=64 and the time step is of the order of 1/20 which corresponds to the time step between two frames in our thermal camera.Nx. j 2 1  k i . in the equation (3.hz is a matrix of size Nx × Nz reshaped from the 2D unknown solution’s array (Ti. j 1 / 2  k i . j  Ti. j 1  Ti. Thx. for the biological tissues...N x i  1. Tni..In the cell centred in (xi.hz   hx . j Ti 1.31). will be showed in the next chapter. j 1 / 2 j  1. j  k i .33 and k i 1 / 2.34 Where. 87 .j). j hz i  1.. Using the boundary conditions on Φ and T we obtain a classic linear system: MT hx ..N z  1 3. j 2     3.32 The heat flux values at the wall of the cells are approximated with :  i 1 / 2.j denotes the temperature at time ndt.. M is pentadiagonal matrix of size (Nx × Nz)2. j  k i .N z 3. i=1. The main thermal and optical costants used in the simulations.zj).hz 3.N x  1  i .z).

results obtained on different patients appling the approach described in the previous chapter will be showed and discussed. a) b) Fig 4. will be proposed. Experimental and numerical values will be analized and compared. Particularly. depth and thickness relative to different plane of vessels have been estimated and in each different clinical situations numerical simulations have been run. b) visible image.1 Introduction In this last chapter it will be shown results obtained through the monitoring of patients affected by plane angioma in the facial region. confirming the great contribute of our methodological approach. 88 . Subsequently. After a description of the instruments used (laser system and infrared camera) and of the operative protocol realized the main results will be presented. in order to emphasize the thermal behaviours associated to differents facial region. some set of laser parameters relative to plane angioma characterized by specific depth and thickness.Chapter 4 EXPERIMENTAL AND NUMERICAL RESULTS 4.1 : Example of laser treatment on patient with plane angioma: a) infrared image. First of all. some preliminaries evaluations relative to temperatures and temporal dynamics found will be illustrated. The chapter is completed with a discussion relative to some limit conditions for laser treatment application on plane angioma pathology. thanks to the pulsed thermography model. Starting from this results.

Deka Dermobeam 2000 Display Laser tool Handle LASER SYSTEM Wavelegth Energy density (Fluence) Pulse duration Spot diameter Emission Laser mode Energy stability in emission 595 nm 3 . the laser treatments on 26 patients affected by plane angioma have been monitored.3.35 J/cm2 0. nose and temple. being haemoglobin the most relevant chromophore in vascular pathologies. as it is possible to see from figure 1. the emission wavelength of this laser 89 . the angioma was widespread in different facial region and principally on cheek. the Dye laser naturally becomes the elective therapeutic tool by selective absorption. The laser therapy is applied in photothermal regime using the selective photothermolysis principle. The laser system used has been Pulsed Dye Laser – Deka Dermobeam 2000 with characteristics showed in fig 4.5 – 40 ms 2 . In the majority of cases.2 : Main characteristics of the Pulsed Dye Laser – Deka Dermobeam 2000 As described in chapter 1. The 26 patients were 17 male and 9 female with age in the range 20-50.2 Materials and method In the experimental activity of this work. In order to study an homogeneous sample group we monitored all patients with a plane angioma in the facial region.2 PDL .7 mm 15 Hz Gaussian 5% Fig 4. In fact.4.5 .

for the Avio camera the acquisition spectral band is within 8-14μm. The main characteristics of the cameras are showed in the fig. The infrared acquisition has been performed by means of two differents digital thermal Camera: a Land FTI6 . In the next pictures the system and laser therapy ambulatory are Fig 4.02 s.3 : Main characteristics of the infrared cameras employed Black body correction has been executed to evaluate and to correct drift and shift effects due to the infrared camera at the beginning of each measurement session.5 μm 0. and the temperature sensitivity is 0. 4.06 °C 14 bit 60 Hz Fig 4.4 : Laser theraphy ambulatory: infrared camera and laser tool 90 . INFRARED CAMERAS Land FTI6 Sensor Spectral range Temperature sensitivity Image resolution Max Frame Rate InGaAs –FPA 256X256 3 .1 °C 8 bit 15 Hz Avio TVS500 Microbolometer –FPA 320X240 8 . Particularly.3.14 μm 0.06 K.95.HgCdTe camera used in preliminaries evaluation and a more performanced Avio TVS 500 – microbolometer camera used in the whole study. the time resolution is 0.(595 nm) is very close to the absorption peak of the target (haemogloblin) as required from selective photothermolysis principle. Emissivity of the skin has been estimated as shown: = 0.

self-implemented software has been realized with a MATLAB platform. 3. starting from referential laser parameters in tab. in the covering of the area of the patient the physician laids a pulse upon the other of about 20 % of spot diameter in order to reduce dischromic effects. each information has been analyzed with the PT model and with the numerical simulation described in chapter 3.1) . The This procedure has permitted the registration of the thermal behaviour of the specific anatomic region under laser-interaction: temperature and its temporal evolution (thermal recovery). Subsequently. 0. during and after (for a time of about 30 s) the laser pulse. In the next sections the results achieved will be showed and discussed starting from some preliminaries evaluations.He defines the laser parameters (usually.5 °C during the measurement.He covers the area chosen with single laser pulses with an interval of 2-4 s Particularly. dust and others .The anatomic facial region of intereset has been cleaned with alcohol in order to remove sweat.He chooses the area of the angioma that he wants to treat (of dimension about 5x5 cm) . the next procedure has been performed sequentially: . cosmetics.The medical protocol commonly used for the treatment of patients with plane angioma pathology by the physician is the following: .Acclimatization of the subject in the therapy rooms for a time not less than 15 minutes . In order to apply the operative approach described in the previous chapter. The laser therapy room has been thermostated at 24 umidity was about 60%.A single pulse with referential laser parameters has been applied in the middle of the region of interest . 91 .An infrared recording of the region under investigation has been made continuously some seconds before. For the data and numerical analysis.

All graphics reported in this section refer to a set of referential laser parameters shown in the next table: Referential laser parameters Wavelenght 595 nm Energy density 7 J/cm2 Pulse duration 1. In this section.3 Preliminary results and observations The first step was aimed to evaluate the thermal responses of different parts of the skin illuminated by laser power. 4.5 ms Spot diameter 7 mm Tab.1 : Set of laser parameters used in the preliminaries monitoring In the first graph temperatures obtained in a specific anatomic region after a single laser pulse are compared for different patients 70 P5 P4 P3 P2 P1 65 60 55 50 45 40 35 0 5 10 Temperature (°C) T Cheek Ti TF 25 °C 15 20 25 30 Event Fig. some graphics of thermal response relative to representatives cases. extrapolated from the monitoring on all patients.4. are reported with the intent of emphasizing the main results obtaind from these preliminaries evaluation. 4.5 : Temperatures obtained after laser pulse refered to cheek region 92 .

In general. 4. obtained immediately after the pulse. The temporal evolutions of the temperature after the laser pulse has been recorded and analyzed also for different anatomics regions.6 which refers to a patient affected by widespread plane angioma. fig. For each patient six thermal response associated to laser parameters in table 4. 93 . from the evaluations performed on all patients it has been found with the same laser parameters. The evaluations refer to the same facial region affected by plane angioma (cheek) for 5 different patients (from P1 to P5). on a same patient and in presence of the same pathology. for a same patient in a same facial region show a good repetitions of the temperatures measured. 4. can be very different. As it is possible to observe. Fig.3 shows how each anatomic region of the face (of a same patients) is characterized from a different recovery time (Tr).6. However. For example. even if the initial temperatures are quite similar for all patients (physiological temperature). 64 60 56 T Cheek Nose Temple 20 °C Temperatura (°C) 52 48 44 40 36 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Event Fig. the final temperature. 4. that the T among differents anatomics regions can be higher then 15-20 °C. both fig.1 are graphicated. this dynamic depends on the region under investigation.In the graph with black and red point we indicated the initial and final temperatures recorded before and after the laser pulse respectively. shows how the temperature obtained immediately after the pulse can change from about 52 °C on the cheek to 64 °C on the tample.5 and 4.6 : Temperatures obtained after laser pulse refered to cheek region Similar situation has been found for the thermal response obtained in different facial region of a same patient affected by plane angioma. In general. 4.

6 Cheek 6.4 0.5 0. In next table the average recovery times evaluated on three facial region are reported Recovery time for differents facial regions Region Tr (s) Temple 2.0 0. 4.7 : Temporal evolutions of the temperature after a laser pulse in three region of the face This dependence has been found in each patient with widespread plane angioma where it has been possible to realize evaluations in different regions affected by pathology.1 0.5 4.5 3.0 3.2 0. 60 56 52 Nose Cheek Temple Temperature (°C) 48 44 40 36 32 -0.0 1.5 2. 4.6 Tab.The recovery time is defined as the time occurring by the tissue for recovering the 70% of the induced thermal gap.0 Time(s) Fig.5 1.2 : Mean values of the recovery time estimated in three different facial regions 94 .0 2.5 Nose 4.

2 refer to mean values obtained on all recovery time estimated in the temporal evolutions recorded. The pulses are numerated from 1 to 4 in temporal order. This is due to the thermal superimposition effect present when the time awaited between two successive pulse is lower than the time spent from the tissue for a complete recovery. emphasise how the thermal behaviours of the tissue achieved under laser treatment for a given set of laser parameteres. depend on the specific target. the temperature induced to the tissue is strictly linked to hystological modification 95 . It is important remember. As it is possible to note all temporal rate reported are quite similar. when the analysis is focused on a same patients in a same anatomic region a good repeatability of the dynamics has been found. As well as for the temperatures. All evaluations shown since here. they can differ from region to region and from patient to patient. Particularly.8: Temporal rate of the temperature refered to the cheek of a patient after application of laser pulses The graph shows four dynamics associated to four single laser pulse made in sequence on four neighbouring points of the tissue. In fig. 4. as it has been discussed in chapter 1 that in phototermal interaction regime.The data in table 4. 68 64 60 Temperature (°C) 56 52 48 44 40 36 0 2 4 4 3 2 1 Time (s) Fig. 4. In the graph.8 temporal rate of the temperature refered to the cheek of a patient and estrapolated from a recorded laser treatment are reported. the initial temperatures refered to the dynamics are different.

These observations justifie and confirm thtat if not using specific and tailored laser parameters for each different clinical situation the laser treatments on plane angioma pathology permits an optimal improvement just in a small number of patients. as discussed in the section 3. Therefore.obtained. object of this thesis work and described in previous chapthers. With this intent the operative approach. becames a necessity. Starting from this observations. changing patient or anatomic region. these preliminary results point out how with the same laser parameters. in order to improve the therapeutic efficiency and the safety in the laser therapy. for each specific treatment. the effects and so the therapeutic efficiency obtained can greatly differ. changing from optimal therapeutic results to unwanted harmful effects (for istance scars formation). 96 .3. the request of optimized set of laser parameters. has been performed and the main results achieved will be shown and discussed in the next section.

In this regions.3: Depth and thickness of the plane of vessels evaluated with the PT model 97 . the recorded dynamics have been analized with the pulsed thermography model descibed in the section 3. Successively.2 has been applied to all patients in the anatomics regions of the face where the angioma was present.3 Thickness (μm) Depth (μm) Thickness (μm) Patient Region Depth (μm) Patient Region 1 2 3 4 5 6 7 8 9 10 11 12 13 cheek temple nose cheek temple cheek nose cheek temple cheek temple nose cheek cheek nose cheek temple cheek temple nose cheek temple nose cheek temple nose cheek cheek temple nose 210 120 150 310 160 170 140 550 520 410 320 330 570 800 670 595 210 350 310 330 265 160 190 430 260 400 470 780 220 470 10 10 10 30 10 10 10 80 80 40 30 30 80 160 100 90 10 30 30 30 20 10 10 40 20 40 40 160 10 40 80 80 70 110 70 70 80 230 150 160 130 150 210 320 340 250 90 310 90 170 160 90 110 220 80 180 140 270 90 250 20 20 20 40 20 20 20 130 60 60 50 60 120 220 240 130 20 210 20 70 60 20 40 120 20 70 50 140 20 130 14 15 16 17 18 19 20 21 22 23 24 25 26 cheek temple nose cheek temple cheek nose cheek temple nose cheek temple cheek temple nose cheek temple cheek cheek temple nose cheek cheek temple nose cheek temple nose cheek nose 660 390 450 630 510 600 540 730 410 480 610 530 590 320 520 900 770 230 520 330 510 420 810 270 720 780 240 710 490 510 100 40 40 100 80 90 80 110 40 40 90 80 90 30 80 180 150 10 80 30 80 40 160 20 140 150 20 140 40 80 240 160 220 340 220 310 180 340 180 250 110 260 180 160 230 370 310 80 240 130 210 160 280 140 330 280 90 310 120 120 130 60 120 240 120 210 70 240 70 130 40 140 70 60 120 250 210 20 130 50 110 60 150 50 230 150 20 210 50 50 Tab. the temporal evolution of the temperature achieved after a laser pulse has been recorded for each clinical situation. The results obtained for each patients are shown in table 4.4. 4. For each patient through this model the depth and thickness of the plane of vessels associated to plane angioma in the region under treatment has been estimated .4.4 Pulsed Thermography model: experimental results The operative procedure described in the section 4.

Despite the previous case. As it is possible to see from the table. 98 . the depth estimated mainly changes in the range from 300 μm to 700 μm while the thickess changes from 100 μm to 300 μm. In the next histograms the values for the depth and thickenss of the angioma are reported without distinction of the anatomic region.9: Distributions relative to the depth and thickness evaluated globally on all anatomic facial region As it is possible to note. the majority of patients were affected from a plane angioma widespread in more region of the face. 50 40 Frequency (%) 30 20 10 0 100 200 300 400 500 600 700 800 900 Depth ( m) 50 40 Frequency (%) 30 20 10 0 0 50 100 150 200 250 300 350 400 Thickness ( m) Fig. 4. in the next histograms the results refer to single anatomic facial region.The evaluations in table 4.3 refer to mean values achieved on 5 different measurement.

99 . for each clinical situation. Subsequently.Area angioma : Naso 50 45 Angioma area : Nose Frequenza percentuale (%) 40 Frequency (%) 40 35 30 25 20 15 10 5 30 20 10 0 100 0 200 300 400 500 600 700 800 900 0 50 100 150 200 250 300 350 400 Profondità ( m) 50 Thickness ( m) Area angioma : Guancia Angioma area : Cheek 40 Frequenza percentuale (%) 200 300 400 500 600 700 800 900 40 35 30 25 20 15 10 5 0 0 50 100 150 200 250 300 350 400 Frequency (%) 30 20 10 0 100 Depth ( m) Area angioma : Fronte Spessore ( m) Area angioma : Fronte Frequenza percentuale (%) 40 Frequenza percentuale (%) 200 300 400 500 600 700 800 900 40 30 30 20 20 10 10 0 100 0 0 50 100 150 200 250 300 350 400 Profondità ( m) Spessore ( m) Fig. From these results. The exerimental results and the simulations refered to differents biological multilayer are reported and discussed in the next section. A higher presence of fat. 4. This is probably due to the higher presence in the cheek of subsurface fat respect to other facial region. probably. a specific morphological reconstruction of the multilayer biological tissue under treatment. In the simulations the thermal behaviour of the biological multilayers has been achieved for different set of laser parameters as input. the deeper and thicker plane angioma has been found in the cheek while the less depth and less thick on the temple.10: Distributions relative to the depth and thickness evaluated on three anatomic facial region As shown in the histograms. These data permit. on these multilayer it has been executed the 2D+1 numerical simulation based on the biometric heat transfer model described in chapter 3. the formation of plane angioma seems to spread preferentially the anatomic region of the cheek. in average. facilitates the extention of the angioma differently from the other regions (for istance the temple) characterized by an higher presence of subsurface muscular tissue.

4.5 Numerical simulations: results

The evaluations obtained in the previous section permit the reconstruction of the biological multilayer made of: epidermis, dermis, plane of vessels and dermis again, associated to different anatomic region of the face for each patient. Subsequently, on these multilayers it has been performed a numerical simulation based on a biometric heat transfer model described in the section 3.5. In the simulations the thermal behaviour of the biological multilayers has been achieved for different set of laser parameters as input. The numerical output consist in the evaluations of the spatials and temporal temperature profiles obtained on the multilayer after a single laser pulse. Next, some results achieved from the simulations will be shown and discussed. Moreover, different clinical situations will be compared among them. Starting from these results, at the end of the section, some set of laser parameters will be proposed for the treament of multilayers with specific plane angioma sizes. The numerical program of calculation has been realized with a home-made Matlab code. In table 4.4 the thermal and optical costants used in the simulations and adviced from literature [110-114] are shown .
Optical and Thermal costants Epidermis Dermis Density Specific heat Conductivity Absorption coefficient Scattering coefficient (g) Refraction index ( =595nm) 1 Kg dm-3 3,7 *10-3 J Kg-1 K-1 0,209 *10-6 W m-1 K-1 18-60 cm-1 470 cm-1 (0.790) 1,37 1 Kg dm-3 3,7 *10-3 J Kg-1 K-1 0,322 *10-6 W m-1 K-1 2,2 cm-1 129 cm-1 (0.790) 1,37

Blood vessels 1,055 Kg dm-3 3,6 *10-3 J Kg-1 K-1 0,492 *10-6 W m-1 K-1 147 cm-1 468 cm-1 (0.995) 1,33

Tab. 4.4: Optical and thermal costants used in the simulations

The optical costants in table 4.4 refer to wavelenght of 595 nm and patients of phototype III. In the calculations the biological multilayer has been divided in a grid of 64x64 cells, the time step between two frames has been fixed to 1/20 s and the depth of the epidermal layer has been fixed to 100 m.

100

- Numerical solution: thermal dynamics associated to the problem

In the next images, a typical case of temporal evolution achieved is shown. The sequences refer to the dynamics simulated on a structure with a plane angioma depth 600 μm and thick 100 μm obtained after a single laser pulse. The laser parameters used as input are reported in the tab. 4.5

laser parameters
Energy density - Fluence Pulse duration Spot diameter 7 J/cm2 5 ms 7 mm

Fig 4.11: multilayer used in the simulation

Tab. 4.5: Laser parameters used in the simulation

0s

1s

2s

3s

4s

5s

6s

7s

8s

9s

10 s

11 s

12 s

13 s

14 s

15 s

Fig 4.12: Temporal sequence of frames extrapolated from the simulations

101

The images reported have been extrapolated from a full sequence achieved from the simulation. The images corresponding to the istant t=0 s refers to the first grid calculated after the laser pulse. As it is possible to note from the sequence the thermal dynamics in the layers isn’t neither homogeneous nor symmetrical as it was expected. For example, the layers more superficial recover quickly than deep layers and the heat, generated from the laser pulse, moves towards the core of the structure. This is mainly due to the different boundary condition present in the model (eq. 3.27) and associated to epidermis (z=0) and dermis (z=D, core) layers. Moreover, as widely discussed, the biological structure is a dishomogeneous multilayer with optical and thermal properties depending from the different tissues. In particular it is possibile to note from fig. 4.12 (at t=0) how the temperature achieved on plane agioma layers is higher than the temperature found for epiderimis or dermis because of the higher absorption coefficient at laser wavelenght (595 nm) of the plane of blood vessels (table 4.4). This particular dynamics has been found in all numerical solutions calculated in different situations (for different multilayers and different laser parameters) and it is peculiar for this problem. Changing the biological multilayers and the laser parameters, this dynamics is obviously characterized from different temperatures for the layers and different recovery times.

102

the temporal evolutions and the spatial distribution of the temperature on the epidermis and on the plane of vessels are compared..Numerical solutions refered to a same structure and for different laser parameters In order to optimize the laser parameters for each specific treatment. Spot diameter = 7 mm tpulse = 5 ms Plane Angioma Epidermis Epidermis . is showed in 2-D and 3-D rappresentation.11 and with the laser parameters of the tab. 4. In the next figures the results obtained from simulation on the biological multilayer in fig.Experimental points 2 Plane Angioma Epidermis Epidermis . the temperature obtained on the plane of vessels and on the other tissues of the structure must be evaluated for different set of laser parameters. Moreover.14: Temporal evolutions and spatial distribution of T on the epidermis and on the plane of vessels 103 . the distribution of the T induced on the structure calcolated at istant t=0 (immediately after the laser pulse).5 are reported.experimental points C 30 C 25 20 15 10 5 0 0 2 4 6 8 10 12 14 0 -8 -6 -4 -2 0 2 4 6 8 10 20 time (s) Length (mm) Fig 4. 2D and 3D rappresetation 45 40 35 30 Fluence = 7 J/cm .13: Distribution of the temperature calcolated at istant t=0 s. 4. In particular. Spot diameter = 7 mm tpulse = 5 ms 40 2 Fluence = 7 J/cm . Fig 4.

14 the temperatures for the plane angioma and the epidermis estimated from the simulation have been rappresented respectively with black and red point.15: Distribution of the temperature calcolated at istant t=0 s. the next graphs show how the numerical solutions change when the spot diameter moves from 7 mm to 2 mm or the pulse duration from 5 ms to 1 ms. Spot diameter = 7 mm tpulse = 1 ms 60 2 Fluence = 7 J/cm . Starting from these laser parameters.Experimental points 2 Plane Angioma Epidermis Epidermis .In the graphs in fig. 4. 2D and 3D rappresetations 60 50 40 30 20 10 0 0 2 4 Fluence = 7 J/cm . The induced thermal gap calculated between the plane angioma and epidermis at istant t=0 s has been ~ 15 °C. In the same graphs.Experimental points C 50 40 30 20 10 0 C 6 8 10 12 14 -8 -6 -4 -2 0 2 4 6 8 time (s) Length (mm) Fig 4. the experimental values recorded on the epidermis with the infrared camera are reported with green points. It is possible to observe as this last are in a good agreement with simulated data.16: Temporal evolutions and spatial distribution of T on the epidermis and on the plane of vessels 104 . Spot diameter = 7 mm tpulse = 1 ms Plane Angioma Epidermis Epidermis . laser parameters Fluence Pulse duration Spot diameter 7 J/cm2 1 ms 7 mm Fig 4.

2D and 3D rappresetations 100 90 80 70 Fluence = 7 J/cm . Spot diameter = 2 mm tpulse = 5 ms Plane Angioma Epidermis Epidermis . at istant t=0 s. a good agreement between numerical solutions and experimental data refered to epidermis layers has been obtained. the difference on the maximum temperature achieved changes from ~ 15 °C to ~ 25 °C when the spot diameter changes from 7 mm to 2 mm and from ~ 15 °C to ~ 45 °C when the pulse duration changes from 5 ms to 1 ms.laser parameters Fluence Pulse duration Spot diameter 7 J/cm2 5 ms 2 mm Fig 4. In fact. both smaller spot diameter and shorter pulse duration induces a thermal gap between epidermis and plane angioma higher than the previous case. In these cases.Experimental points 2 80 C 60 50 40 60 40 30 20 10 0 0 2 4 6 8 10 12 14 0 -8 -6 -4 -2 0 2 4 6 8 20 time (s) Length (mm) Fig 4.experimental points C 2 100 Fluence = 7 J/cm .18: Temporal evolutions and spatial distribution of T on the epidermis and on the plane of vessels These last graphs show how.17: Distribution of the temperature calcolated at istant t=0 s. Spot diameter = 2 mm tpulse = 5 ms Plane Angioma Epidermis Epidermis . 105 .

20: Temporal evolutions of T on the plane of vessels fordifferent fluence 106 . tpulse = 5 ms Fluence = 4 J/cm 2 2 2 50 Fluence = 7 J/cm Fluence = 10 J/cm 40 C 30 20 10 0 0 2 4 6 8 10 12 14 time (s) Fig 4. laser parameters Pulse duration Spot diameter 5 ms 2 mm 4 J/cm2 7 J/cm2 10 J/cm2 Fig 4.In the next numerical results the dynamics of the thermal gap induced by the laser pulse on the biological structure of fig. A 2-D and a 3-D rappresentation is given in all three different situations. In the first case. 4. the pulse duration and spot diameter have been fixed while three different values for the fluence have been used. 2D and 3D rappresetations Plane Angioma 60 Spot diameter = 7 mm .11 are compared when two of the three laser parameters are fixed and just one is changed.19: Distribution of the temperature calcolated at istant t=0 s.

the fluence and spot diameter have been fixed while two different values of the pulse duration have been used.22: Temporal evolutions of T on the plane of vessels fordifferent pulse duration 107 . 4. Spot diameter = 7 tpulse= 1 ms tpulse= 5 ms 2 C 25 20 15 10 5 0 0 2 4 6 8 10 12 14 time (s) Fig 4. laser parameters Fluence Spot diameter 4 J/cm2 2 mm Fig 4. 2D and 3D rappresetations Plane Angioma 45 40 35 30 Fluence = 4 J/cm .21: Distribution of the temperature calcolated at istant t=0 s.20 shows clearly how the induced thermal gap can change from ~ 25 °C to ~ 60 °C for the range of fluence used in these simulations (4-10 J).The graph in fig. In the second case.

tpulse= 5 ms 80 70 60 50 40 30 20 10 0 0 2 4 6 8 10 12 14 2 Spot diameter = 2 mm Spot diameter = 7 mm C time (s) Fig 4. in the calculations. two values for the spot diameters have been used while the others two parameteres have been fixed laser parameters Fluence 7 J/cm2 Pulse duration 5 ms Fig 4.23: Distribution of the temperature calcolated at istant t=0 s.24: Temporal evolutions of T on the plane of vessels fordifferent pulse duration 108 .In the last case. 2D and 3D rappresetations Plane Angioma Fluence = 4 J/cm .

109 . All simulations and results shown in this section refer to the laser parameters reported in the table 4. . 2D rappresentation. Fig 4.Numerical solutions refered to different structures with the same laser parameters In this paragraph the results relative to the biological multilayer with a plane angioma at different depths or with different thicknesses are shown. 600 μm and 900 μm while the thickness was 100 μm.25: multilayer used in the simulations Fig 4. The depth of the angioma inside the three structures was respectively 300 μm . The first case is relative to three structures with a plane of vessels at different depth but with the same thickess.As it is possibile to observe from the last graph.5. the induced thermal gap from the laser pulse can vary massively when the spot diameter is changed. These structures associated to three different clinical situations have been numerically analysed.26: Distribution of the temperature calcolated at istant t=0 s.

this trend restricts just to patients affected by plane 110 . in case of both epidermis and plane angioma layers. 4 the temporal evolution and the spatial distribution of T. As it is expected. From the images in fig. For epidermis. Spot diameter = 7 mm tpulse = 5 ms Thikness = 100 m 2 Plane angioma 100 Fluence = 7 J/cm . Spot diameter = 7 mm tpulse = 5 ms Thikness = 100 m Depth = 900 m Depth = 600 m Depth = 300 m 2 Epidermis 35 30 25 20 Fluence = 7 J/cm . with a gap of ~17°C. Spot diameter = 7 mm tpulse = 5 ms Thikness = 100 m Depth = 900 m Depth = 600 m Depth = 300 m 2 C 20 15 10 5 0 0 2 4 6 8 C 15 10 5 0 10 12 14 -8 -6 -4 -2 0 2 4 6 8 time (s) Length (mm) Plane Angioma 80 Fluence = 7 J/cm . As it will be discussed later.27: Temporal evolutions and spatial distribution of T on the epidermis and on the plane of vessels As in the previous case the false colours images extrapolated from the numerical solutions refer to the istant subsequent to the laser pulse (t=0 s). are higher on the plane angioma layers than epidermis layers. the temperatures obtained on the plane angioma are higher than the temperatures on the epidermis but the ratio Tangioma / Tepidermis decreases for deeper plane of vessels. The simulations show how the difference found on the temperatures on structures affected by plane angioma to different depth in the tissue. and for plane angioma layers from ~25 °C to ~75 °C. Spot diameter = 7 mm tpulse = 5 ms Thikness = 100 m Depth = 900 m Depth = 600 m Depth = 300 m 2 60 80 40 Depth = 900 m Depth = 600 m Depth = 300 m 60 C C 40 20 0 20 0 0 2 4 6 8 10 12 14 -8 -6 -4 -2 0 2 4 6 8 time (s) Length (mm) Fig 4.Epidermis 40 35 30 25 Fluence = 7 J/cm . T change in the range from ~23 °C to ~ 40 °C. with a gap of ~50°C. are graphicated.

angioma which are not too deep.28: Distribution of the temperature calcolated at istant t=0 s. the numerical solutions relative to two biological structures with a plane angioma at same depth but with different thicknesses have been compared. while the depth was 600 μm. In the next case analized. In this case a laser therapy can supply good results and reduced risks. Fig 4. 2D rappresentation 111 . For the two structures the thickness was 100 μm and 300 μm respectively.

Spot diameter = 7 mm tpulse = 5 ms Depth = 600 Thikness = 300 m Thikness = 100 m 2 Depth = 600 m Thikness = 100 m Thikness = 300 m 35 30 25 C 20 15 10 5 0 0 2 4 6 8 C 10 12 14 20 15 10 5 0 -8 -6 -4 -2 0 2 4 6 8 time (s) Length (mm) Plane Angioma Fluence = 7 J/cm . Spot diameter = 7 mm tpulse = 5 ms Depth = 600 Thikness = 300 m Thikness = 100 m 2 50 50 40 C C 0 2 4 6 8 10 12 14 30 Thikness = 100 m Thikness = 300 m 40 30 20 20 10 10 0 0 -8 -6 -4 -2 0 2 4 6 8 time (s) Length (mm) Fig 4.29 the variations on the temperatures achieved in the two situations for the plane angioma and for epidermis are. 112 .29: Temporal evolutions and spatial distribution of T on the epidermis and on the plane of vessels Differently from the previous case. 4. the same. These results demonstrate how the thermal behaviour achieved in the multilayer biologial structures affected by plane angioma is more sensitive to the variations of the depth than variations of the thickness. Spot diameter = 7 mm tpulse = 5 ms 40 2 Epidermis Fluence = 7 J/cm . from the graphs reported in fig. substantially. The data refer to evaluations achieved in the simulations. with a gap in both situations of about ~9-10°C. In the next graphs the ricovery time and the induced thermal gap ( T) versus the depth of the plane angioma in the tissue are reported. In fact. the maximum ~42 °C to ~52 °C while the maximum T found for the plane angioma goes from T found for the epidermis goes from ~ 28 °C to ~37 °C.Epidermis 35 30 25 Fluence = 7 J/cm . Spot diameter = 7 mm tpulse = 5 ms Depth = 600 m 2 Plane Angioma 60 Fluence = 7 J/cm . The values are relative to plane angioma with thickness 100 μm and 300 μm and various depths.

In the graphs the experimental values refered to epidermis and extrapolated from some patients monitored are reported with green points. the T decreases with depth and increases with the thickness of the plane of vessels. The graphs show clearly a good agreement between simulated data and experimental values in both situations. it’s clear.31: Induced thermal gap versus depth for two different thickness From the graphs.30: Recovery time versus depth for two different thickness Plane Angioma 100 90 80 70 Fluence = 7 J/cm .Thickness = 100 m Thickness = 300 m Thickness = 100 m . Differently from the previous case. Spot diameter = 7 mm tpulse = 5 ms 2 Plane Angioma Ricovery time (s) 5 Epidermis 4 3 2 1 300 400 500 600 700 800 900 Depth ( m) Fig 4.Experimental value 7 6 Fluence = 7 J/cm . 113 . how the ricovery time increases with the depth while it doesn’t change at all to variations of the thickness. Spot diameter = 7 mm tpulse = 5 ms Thickness = 100 m Thickness = 300 m Experimental points 2 C 60 Epidermis 50 40 30 20 300 400 500 600 700 800 900 Depth ( m) Fig 4.

6 Absorption coefficient Phototype I Phototype III Phototype IV 7 J/cm2 5 ms 7 mm Tab. Spot diameter = 7 mm tpulse = 5 ms Phototype VI Plane Angioma Epidermis 2 C 30 25 20 15 10 5 0 0 2 4 6 8 10 12 14 Plane angioma Epidermis Plane Angioma Epidermis 0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14 time (s) time (s) time (s) Fig 4. the thermal effects achieved on different phototypes have been compared.33: Temporal evolutions of T on the epidermis and on the plane of vessels for different phototypes 114 . In this study. 4. Spot diameter = 7 mm tpulse = 5 ms Phototype I C 2 55 50 45 40 35 30 25 20 15 10 5 0 Fluence = 7 J/cm .33. The absorption coefficients for the epidermis used in the simulations are reported in table 4.11) and laser parameters reported in table 4. 4.6: Absorption coefficients for the epidermis used in the simulations Ph I Ph III Ph VI Fig 4. In fig. The results refer to structures with a plane angioma depth 600 μm and thick 100 μm (fig. 4..Numerical simulation for different phototypes All numerical simulations and results up to now reported refer to patients of phototype III (typology more frequent). Spot diameter = 7 mm tpulse = 5 ms Phototype III C 2 50 45 40 35 30 25 20 15 10 5 0 Fluence = 7 J/cm . patients of different phototype are characterized from a different absorption coefficient for the epidermal layers.32: Distribution of the temperature calcolated at istant t=0 s for different phototype Phototype VI 55 50 45 40 35 55 Fluence = 7 J/cm .5. numerical simulations on three equal multilayers (with same dimensions for the plane of vessels) but with different phototypes are shown. From an optical point of view.

clearly.In figure 4. In next figure the temporal rate of T refered to the epidermis and plane angioma are reported separately for the three phototypes 55 55 Pht VI 50 45 40 35 Fluence = 7 J/cm . Spot diameter = 7 mm tpulse = 5 ms C 2 50 45 40 35 Pht I Pht III Pht VI Fluence = 7 J/cm .33 the temporal rate of the T calculated on epidermis and on plane angioma has been compared for each different phototype. the thermal gap between epidermal layer and plane angioma layer increases going from phototype VI to phototype I and in particular. For the epidermis.34: Temporal evolutions of T on the epidermis and on the plane of vessels for different phototypes The graphs show. 115 . From the graphs. for phototype VI the laser pulse induces a maximum T on the epidermis (for t= 0 s) higher than the maximum T obtained on the angioma. Spot diameter = 7 mm tpulse = 5 ms Plane angioma Phototype VI Phototype III Phototype I 2 C 30 25 20 15 10 5 0 -2 Pht III Epidermis Phototype VI Phototype III Phototype I 30 25 Pht I 20 15 10 5 0 0 2 4 6 8 10 12 14 -2 0 2 4 6 8 10 12 14 time (s) time (s) Fig 4. as going from phototype I to phototype VI that the temperature obtained on the epidermis increases while the temperatures found on the angioma decreases. the maximimum gap of the temperature has been estimated to be about ~37 °C (from ~18°C for phototype I to ~55°C for phototype VI) while for the angioma it has been found of ~13 °C (from ~37 °C for phototype VI to ~50 °C for phototype I).

in the same time. in the laser therapy performed in the photothermal regime. then for the tissues of phototype V-VI (where SI therapy can be hazardous. as the same situation (SI 1) is found for angiomas deeper than about ~900 1) the application of laser m (as it has seen above) independently from the phototypes.4. the temperatures induced on the tissues are strictly linked to the histological modification obtained. It means. TDERMIS / TANGIOMA > 1 Fig 4. if the ratio SI = Tepidermis / Tangioma can be regarded as a selectivity index of the interaction. and. 116 . the results achieved by simulations advises against the application of the laser treatments. in these case too.33 and 4. In general. In this patient the lower temperatures achieved on the angioma makes the blood vessels photocoagulation improbable and at the same time the higher temperatures achieved on the epidermis and dermis can produce the formation of scars due to the collagene denaturation. The last results in fig. the goal of the laser therapy is to obtain a local and selective interaction with the target structures (selective photothermolysis principle). a lower interaction with the surronding tissues (dermal and epidermal layers) is desired. to maximize the temperature on the plane angioma and minimize the temperatures on the epidermis and dermis induced by a laser pulse. for treatment on plane angioma.6 Limit conditions for applications of laser treatments As discussed in chapter 1. In fact.34 emphasize how in some circumstance this is impossible. pratically. 4. It is possible to note. so.35: Distribution of the temperature calcolated at istant t=0 s for a phototype III with a plane agioma depth 900 m and thick 100 m. Particularly. a complete photocoagulation of the blood vessels and.

give a wide vision about the the thermal behaviour of the biological multilayer affected by plane angioma when they are subject to selective laser interactions. The set shown refers to patients of phototype III. in tab. as an output of this work it comes that it is possible to choose a set of laser parameters in order to accomplish these relations for each different biological multilayer TEPIDERMIS.4. are chosen in oder to photocoagulate the blood vessels of the angioma and at the same time to reduce the risks of scars formations. These parameters are given with the aim to achieve a more specific and appropriate interaction with the tissue of the patient and to make laser treatments more safety and more efficient. 117 . different set of laser parameters can be proposed.7 Final conclusions: set of laser parameters The results shown up to now together with the complete study on the patients. others set can be suggested for different phototype and for different sizes of the plane angioma. 45 °C < TDERMIS < 30°C Safety Photocoagulation range TANGIOMA < 75 °C The above ranges. These results represent a first step in order to optimize laser parameters for specific clinical situations. Just as an example. 4. According to these indications. In fact. With this approach.7 some set of laser parameters are proposed for treatment of patients affected by plane angioma of different dimensions.

7: Some set of laser parameters proposed for a phototype III and for differents dimesions of the plane angioma 118 .Thickness Depth 100 μm 300 μm 300 μm Fluence Spot diameter Pulse duration = = = 3 J/cm2 7 mm 5 ms Fluence Spot diameter Pulse duration = = = 3 J/cm2 7 mm 7 ms 600 μm Fluence = Spot diameter = Pulse duration = 9 J/cm2 7 mm 5 ms Fluence = Spot diameter = Pulse duration = 5 J/cm2 7 mm 5 ms 800 μm Fluence = Spot diameter = Pulse duration = 16 J/cm2 7 mm 5 ms Fluence = Spot diameter = Pulse duration = 11 J/cm2 7 mm 5 ms Tab 4.

In case of depeer angioma (d > 900 m). the set proposed will be gradually tested on different targets with the aim to obtain. Therefore. for more superficial angioma. subsequently. particularly. with the approach described a good agreement between numerical solutions and experimental results has been obtained for both ricovery time and temporal evolution of T in the majority of cases. eventually. in the next steps of this research activity. a mismatch between numerical solution and experimental values have been often found due to a higher error achieved in the evaluation of the dimensional parameters of the angioma with the PT model used. a complete validation of this novel approach. new PT models will be tested in order to achieve a more accurate evaluation of the dimensional parameters of the plane angioma and. the results up to now obtained in this study are encouraging and a further goal is to extend the analysis from a relatively small number of patients monitored to a larger population. 119 .In conclusion. However.

and to the selective photothermolysis principle which are core of the present work. Different definitions of thermal contrasts used for inversion algorithms have been given and compared: contrast for the transmission mode. of the methodological approach commonly used and protocols. Moreover. lock-in thermography and vibrothermography. The state of art of the laser application in the medical field has been made in order to have an overview of the different use of lasers in medicine. Subsequently. the photoablation. Greatest attention has been given to the description of the photothermal regime of interaction. that differ one another mainly in the way data is acquired and/or processed. the main laser-tissue interaction mechanisms as the photothermal interaction. At the same time the results and limits of the therapy have been estimated. of the laser systems applied. An overview on the infrared detector as bolometers. In particular. have been reviewed and discussed: pulsed approach. the novel approach and the new applications of infrared imaging in medical science has been discussed. quantum detectors.General Conclusions In this thesis the laser – biological tissue interaction has been studied and analized with the support of the infrared imaging technique. Both passive and active approach have been explained. This work started with an overview on the fundamental theoretical concepts of the laser–tissue interaction. step heating or long pulse method. the photochemical interaction. pyroelectric detectors. mainly focusing pricipally the attention on the different active procedures. to the thermodynamic effect. Several active thermographic techniques commonly used. 120 . Focal Plane Arrays (FPA) devices and on the their main performance parameters has been made. contrast for the reflection mode and the differential absolute contrast (DAC). In this description more emphasis has been given to the presentation of the pulsed termography method because it is a part of the experimetal procedure used in this activity. the principles and applications of the infrared imaging technique have been reported. the plasma-induced ablation and the photodisruption have been described. The problems relative to the atmospheric trasmittance and infrared sensors have been treated.

In the third chapter. the core of the research activity has been described. firstly. In particular. Particularly. In general. 4) The data recorded have been processed with the pulsed thermography model and then with the numerical simulation based on a biometric heat transfer model according to the operative approach described in the chapter 3. The experimental activity was constituted of the following steps: 1) Assembly of the monitoring station including a thermal camera. First of all. with a T 25 °C too. the pulsed thermography model and the numerical approach adopted in the operative protocol performed have been explained rigorously. a data acquisition system and a laptop to elaborate the IR images collected. A Similar situation has been found for the thermal response obtained in different facial region of a same patient affected by plane angioma. achieved immediately after the pulse. The research activity started. All patients have been monitored along a cycle of therapies of about 18 months. The novel approach proposed for the laser parameters optimization in the plane angioma treatment has been illustrated and explained in details. The pathology analized (plane angioma) and computation tools used have been treated and discussed. During this time different data refering to the thermal behaviour of the tissue under treatment have been recorded. 2) Introduction of the monitoring station in the operative unit of photodynamic and phototherapy ambulatory of the University of Naples "II Policlinico". 3) From the results of the preliminaries test a standard procedure for data acquisition has been conceived (as described in section 4. can be very different.2). it has been shown how comparing the thermal response obtained in a specific anatomical region after a single laser pulse on different patients (with the same laser parameters). the results relative to preliminaries test and observations have been illustrated. 121 . the final temperature. with a series of preliminaries clinical tests on patients affected by Plane Angioma. The main experimental and numerical results found with this approach have been reported and discussed in detail.

However. changing patient or anatomical region. when the analysis is focused on a same patient in a same anatomical region a good repeatability of the dynamic has been achieved. On these multilayer the 2D+1 numerical simulation has been executed. Subsequently. these results point out that with the same laser parameters. the T among different anatomical regions can be higher than 15-20 °C. the effects and so the therapeutic efficiency obtained can differ a lot. Particularly. this dynamic depends on the anatomical region under investigation and different average recovery times have been evaluated for each facial region. Moreover the deeper and thicker plane angioma has been found in the cheek region as an average. changing from optimal therapeutic results to unwanted harmful effects (for istance scars formation). in the phototermal interaction regime. The temporal evolution of the temperatures after the laser pulse has been also recorded and analyzed for different anatomical regions. they can differ from region to region and from patient to patient. As reported. depend on the specific target. Therefore. In general. as well as for the temperatures. The depth estimated mainly changes in the range from 300 μm to 700 μm while the thickess changes from 100 μm to 300 μm. These evaluations allowed a specific morphological reconstruction of the multilayer biological tissue under treatment for each clinical situation. for a same patient in a same facial region it has been obtained a good repetitions of the temperatures measured. For each patient the depth and thickness associated to plane angioma in the specific region under treatment evaluated with the PT model has been showed. the results obtained using the novel approach proposed have been illustrated. These observations justify and confirm if we don’t use specifics and tailored laser parameters for each different clinical situation the laser treatments on plane angioma pathology permits an optimal improvement just in a small part of patients. 122 . on a same patient and in presence of the same pathology. However. the temperature induced in the tissue is strictly linked to the hystological modification obtained. The numerical program of calculation has been realized with a “home-made” Matlab code. while the less depth and less thick has been found on the temple region.from the evaluations performed on all patients it has been found how with the same laser parameters. All evaluations obtained in the preliminary tests. emphasise how the thermal behaviours of the tissues achieved under laser treatment for a given set of laser parameteres.

for more superficial angioma. on the multilayer biologial structures affected by plane angioma. 600 μm and 300 μm) but with the same thickess (100 μm) have been reported. the thermal effects achieved on different phototypes have been compared. for phototype VI the laser pulse induces a maximum T on the epidermis (for t= 0 s) higher than the maximum T obtained on the angioma. Differently from the first case. above all. In particular. and moves from ~25 °C to ~75 °C (with a gap of ~50°C) for plane angioma layers. In a first case. T moves from ~23 °C to ~ 40 °C (with a gap of ~17°C) for epidermis. A typical case of temporal sequence made of grid representing the temperature distribution and calculated with the numerical model has been shown and discussed. the variations on the temperatures achieved for the plane angioma and for epidermis has been. The temporal evolutions and the spatial distribution of the temperature on the biological layers have been compared. with a gap in both situations of about ~9-10°C. The induced thermal gap ( T) calculated between the plane angioma and epidermis at istant t=0 s refering to the reference laser parameters has been found ~ 15 °C. In a second case. In this situations. numerical solutions refering to different structures with the same laser parameters have been compared. is more sensitive to the variations of the depth than variations of the thickness. when deeper angioma are taken. Changing the spot diameter or the pulse duration the induced T can increase up to ~ 25 °C and up to ~ 45 °C respectively. This last evaluation emphasizes how the thermal gap between epidermal layer and plane angioma layer decreases going from phototype I to phototype VI. At a second stage. These results demonstrate how the thermal behaviour achieved under treatment. In all situations the experimental values and the numerical results have been compared and. in the majority of cases. Subsequently. 123 . numerical solutions refered to a same structure and for different laser parameters have been illustrated.In the simulations the thermal behaviour of the biological multilayers has been achieved for different set of laser parameters as input. the results refering to three structures with a plane of blood vessels at different depth (900 μm. two biological structures with a plane angioma at same depth (600 μm) but with different thickness (100 μm and 300 μm) have been compared. the same. In fact. a good agreement has been obtained for both recovery time and temporal evolution of T. Next. the maximum T found for the plane angioma goes from ~42 °C to ~52 °C while the maximum T found for the epidermis goes from ~ 28 °C to ~37 °C. substantially.

. From the evaluation realized. The output of this research could supply the doctor with very clear indications.After these results. Our further aim is to obtain a complete validation of this novel approach on this research activity will be done according to the next steps: - Testing new PT model in order to achieve a more accurate estimation of the dimensional parameters for the plane angioma. focusing on particular situation (higher phototype or deeper angioma) where the results achieved advise against the application of the laser treatments.Prototipal realization of a infrared imaging system to support the medical laser tool. according to the patient. it is possible to choose a set of laser parameters for each different biological multilayer in order to optimize the thermal interaction and so to make laser treatments safer and more efficient. . The final output will be a methodic scheme which could be employed by the operator no longer in an empirical way but now based on specific patient and phototype skin.Testing this approach for different patologies and in different medical fields. In conclusion in this thesis it has been demonstrated that functional IR imaging can be used with remarkable advantages in monitoring the physical action of the laser ligth thus giving to the operator (the physician) information to optimize his activity. The monitoring system which will be designed in the future in these activities could be of interest for medical industries which operate in this sector on a national and international scale. on how to use the laser in terms of power and duration of the treatment. in terms of both therapy effect and safety. .Gradual application of the proposed laser parameters set on different targets. the limit conditions for application of laser treatments have been discussed. .Validation of the operative protocol proposed. 124 . Some set of laser parameters have been proposed and illustrated for treatment of patients affected by plane angioma of different dimensions.

“Ocular laser threshold investigations”. “Studies of thermal injury”. Henriques. Weinman. pp. pp. J. J. Am. IEEE J. “Laser-Tissue Interactions“. ch.S. L. Tech. Electron. Aerospace Med”. “A new look at multiple scattering”. IEEE J. pp. “On the theory of electron multiplications in crystals”. J. Report NASA. “Avalanche ionization in transparent dielectrics induced by intense light pulses”. 489-502 (1947). ch. 1845-1852 (1987).G. Aerospace Med. Schoenlein. Christian. QE-23. [8] [9] [10] W.J. 33. Zaneveld. N.F. L. “Computer simulations for the irradiating planning of LITT”. Seitz. 1481-1489 (1984). Sci. Molchanov. Bille. Lorenz. “Acoustic studies of the role of immersion in plasma-mediated laser ablation”. van de Hulst. F41609-70-C0002 (1971). Niemz. Tech. “The optics of the human skin”. [14] P. SAM-TR. R. J. pp. Dedrick. Vassiliadis. Rev. 2452-2459 (1976).H. J.A.C. Stern. C. “Diffuse radiation in the galaxy”.W. B. “Optoelectronics and Photonics: Principles and Practices”. Lasers Surg. 125 . Richter. 93.O. pp. 107. “Plasma mediated ablation of corneal tissue at 1053 nm using a Nd:YLF oscillator/regenerative mplifier laser”. Electron. J. [12] F. Nishioka. pp.O. picosecond and femtosecond laser at 532 nm and 625 nm”. W.H. Anderson.J. Joseph. Arch. T. 70. Wiscombe. N.F.G. 23.References [1] R. Pathol. [13] A. Dobi. Rep. Med. Soc. John Wiley Interscience. J. 18-24 (1993). Fujimoto. Birngruber.G. pp. J.R. Springer Berlin Heidelberg New York. Anderson. 70-83 (1941).13 (1981). 587-592 (1989). 426-431 (1991). Kasap. Puliafito. Yariv.C. Reynolds. Teng. W. H. [11] A. QE – 20. Med. A. Invest Dermatol 77. G. Phys. [17] F. Parrish. [16] M.S.N. K. Weinberg.T. [3] [4] [5] [6] [7] S. E. 1 (2003). Henvey. R. 76. 4. 749.1 (2001). (1962).C. Birngruber. Qu. Roggan. “Approximate two parameter phase function for light scattering”. Niemz.R. [2] M. “The change in light reflection of the retina during therapeutic laser-photocoagulation”. [15] D. Klancnik.A. Phys.pp. H. “Corneal ablation by nanosecond. E. J. A. Atmos.H. Prentice-Hall. “The delta-Eddington approximation for radiative flux transfer”. New York. “Quantum Electronics”. Takata. 77-38 (1977). J. 1 (1975). Sov. Opt. Qu. pp. L. 1206-1212 (1980). McCormick. pp. R. “Laser induced thermal damage in skin. Greenstein. 1376-93 (1949). J. Astrophys.G. Rep. ch. pp. Muller. 11. R. (1970).A. Deutsch. [18] A. Ophthalmol. Solid State 12.

L. “Treatment of childen with port-wine stains using the flashlamp-pulsed tunable dye laser”. T. Muller. 287–292 (2000). Science 220. [28] W. J. Kimel. Steffen. 75. Landsberg (1989). Gange. 29. Anderson.G. Wheeland. G. 6. [20] N. J.S. “Use of the Nd:YAG laser to open the posterior capsule after lens implant surgery: a preliminary report”. 320. R. Med. J.R. D.H. Steinert. [27] M. pp.J. Opt. “Wavelengths for laser treatment of port wine stains and teleangectasia”. pp. N.W. “Clinical studies on the efficiency of high power laser radiation upon some structures of the anterior segment of the eye”. “Theory of electron-avalanche ionization induced in solids by electromagnetic waves”. R. Anderson. 1442-1448 (1984). Kenny. Epifanov. Welch. SPIE 2077.[19] N.A. “Modeling of Blood Vessel Heating by Laser Radiation”. Sacchi. “Short-pulsed Nd:YAG laser microsurgery of the eye: biophysical considerations”.R. J. Welch. Phys.S. J. Aron. pp. Laser Surg. R. M. 1091-1098 (1990). G. Ecomed – Verlag. 524-527 (1983). J. [30] R. pp. Aron Rosa.T. QE-10. Lett. Phys. pp. A. J. A. “Laser in Dermatologie. [25] D.J. 674-678 (1973). pp. R.A. Flotte. Scheibner. [24] M. 126. QE-20. 216-226 (1994). Shell. Michauud. R. Zheltov. Sherwood. “Selective photothermolysis: precise microsurgery by selective absorption of pulsed radiation”. van Gemert. Tan.J. eds. “Avalanche ionization and the limiting parameter of filaments induced by light pulses in transparent media”. pp.W. “Laser photothermolysis of single blood vessels in the chick chorioallantoic membrane (CAM)”. Laser Surg. P. “Laser-induced electric breakdown in water”. Qu. pp. 337-45 (1991). Bloembergen. Surg. J. Med. Astafeva. [29] O. Med.A.C. [21] A. Van der Zypen. [34] A. [26] F. Implany. Soc. Seipp. pp. E.T. O. Thyzel. J. Rubanov. Gonzalez. pp. Electron. 129-139 (1981).J. IEEE J. M.J. J.C. Parrish. White. 2. Griesemann. Qu Electron. 907-910 (1972). 352-354 (1980). 416-421 (1989). Gijsberg. N. Tan. “Laser induced electric breakdown in solids”. Berlien. Arch. pp. Pickering. 16. A. Taylor. Berns. Dover. T. K. Ophthalmol. Yablonovitch. Frankhauser. “Laserpuncture of anterior chamber angle in glaucoma”. G. 2018-2022 (1981). pp. pp. W. 147-155 (1995). [31] S. Dermatol. 893-899 (1990). Rev. Seipp. “A superior method of tatoo removal using the Q-switched ruby laser”. Angewandte Lasermedizin“.I. Am. 126 . H. [35] C. Chrenkova. “Treatment of tatoos by Q-switched ruby laser. “Time costants in thermal laser medicine”. IEEE J. Puliafito. A dose response study”. Hammer-Wilson. V. [23] C. G. M. J. Intraocul. Nelson.W. 16. Qu. 90.E Milner. Proc. Engl. QE-17. Soc. Am. [22] C.P.S.O. Gilchrest. B. Qu Electron. pp. Bloembergen. A. Krasnov. N. 9.J. E. Oncol. J. Haina.R. Ophthalmol. IEEE J.M. G. Am. B8. Ing. van Gemert. pp. 375-386 (1974). Dermatol.S. A. Roussel. Int. [33] M. Svaasand. [32] L.F. 3. Opt. pp. 405-421 (1989).

Ljungberg. J. Niemz. Lappe. 5. J. Marinetti. Chen. pp. [37] M. Grinzato. pp. Springer-Verlag. 84-91 (1998). SPIE TT13. Kaplan. “Moisture evaluation by dynamic thermography data modelling”. [38] W. C.[36] M. H. 211-252 (1994). T. “Thermal infrared non destructive evaluation of moisture content in building” CNR Congress on Moisture. SPIE. Grinzato. R. [46] S. Infrared Physics & Technology 49. Rössler. [47] E. Shepard.P. Bison.F. [42] X. Maldague. P. D. pp. [40] M. Hofstädter.H. F. S. E. 211–218 (2002).P. V.P. Proc. J. Study Cases from Northern Italy”. J. Bille. Orlando (USA) (1994). Bison. Dec. L. Nicolai. SPIE 2245. Varenna (Italy) (1994). T. Orlando (USA). SPIE 3255. Proc. H. pp. Dent. [43] H. X.Marinetti. [48] E. “Intrastromal ablations for refractive corneal surgery using picosencd infrared laser pulse“. Bujas-Dimitrijevic. Infrared Phys Technol 43. 347-357 (1999). green and IR picosecond laser pulse”. [44] H. Robba. 127 . “Heating of industrial sewing machine needles: FEA model and verification using IR radiometry”. F. John Wiley & Sons. “Process control using IR sensors and scanners”. Wieland. J. W.G. New York (2001). 74. “Infrared techniques in building and structures: operation and maintenance”. LASERmed 93. diagnostic applications. Fischer. (1993). “Practical applications of infrared thermal sensing and imaging equipment”. Liasi. Proc.V. “Ultrashort laser pulse in dentistry. SPIE 2079. Snell. “Thermographic inspection of TBC coated gas turbine blades: Discrimination between coating over-thicknesses and adhesion defects”.H. Lemieux.F. pp. Laser light Opthalm. International Advances in Nondestructive Testing”. pp. [45] S. New York (1993) [39] M.F.V. “Theory and Practice of Infrared Technology for Nondestructive Testing”.P.P. Proceedings of SPIE. S. Thermosense XVI. 1194-1199 (1995). “Advances in Pulsed Thermography”. 511–515 (2001).Ludwig. Kaplan. G. “Surface ablation of corneal tissue using UV. Sakagami. Kubo. Heidelberg. [50] Q. pp. J. Cernuschi. Niemz. Bison. “Optimal conditions to detect moisture in ancient buildings.M. G.Å. 131-139 (1994). Vavilov. F. Maldague. Niemz. Du. Juhasz. Thermosense XX-An International Conference in thermal sensing and imaging.H. Proc. Rosina. eds. Thermosense XXI. H. D. Loesel. Bille. Rosi. pp. Snell. Niemz. Proc. S. Monograph Series. 92-95 (1997). Berlin. Ed. Marinetti. J. pp. [49] E. “Applications of pulse heating thermography and lock-in thermography to quantitative nondestructive evaluations”. [51] S. Spring. H. A. Hoppeler. Thermosense XXIII. Grinzato. “Infrared Methodology and Technology. pp. 137. 3700. Orlando (USA) (1998).R. “Infrared thermography”. The Engine Yearbook (2004). M. 281-285 (2007). “Cavity preparation with the Nd :YLF picosecond laser”.advantages and limitations”. Photonics Spectra. Li. E. N. Res. [52] T. 149-155 (1993). [41] R.

Technol.P. pp. J. 47. Appl. S. pp. Lewak. 45. Research techniques in nondestructive testing. [56] A. pp.O.E.P.P. 253-262 (1988). Degiovanni. Lamontagne. pp. “Non-destructive thermal evaluation of delaminations in a laminate: I.P. pp. A. J. 43.S. Ziadi. [55] X. “Theoretical and practical aspects of the thermal nondestructive testing of boned structures”. Bendada. Batsale. X. Largoutit. Ludwig. 14-27 (1996). Rev. A. Maillet. 3-23 (2007). Carlslaw. “A combinated integral transform asymptotic expansion method for the characterization of interface flaws through pulsed infrared thermography”. 617-621 (2006). [59] D. [58] D. Lamine. “Advances in pulsed phase thermography”. M. 2694. pp. A. experiments”. P. 5. C. “Heat losses and 3D diffusion phenomena for defect sizing procedures in video pulse thermography”. Gen. [68] A. [66] H. 238-279 (1982).V. pp. F. J.P. 79. “Pulse phase infrared thermography”. 7A. NDT&E Int. Erchiqui. Inverse Problems 21. “Pulsed thermography in the evaluation of an aircraft composite using 3D thermal quadrupoles and mathematical perturbations”. Identification by measurement of thermal contrast”. pp. “Reconstruction of a non uniform interface thermal resistance by inverse conduction”. NDT and E Internat. Bendada. “A study of defect depth using neural networks in pulsed phase thermography: modelling. Maldague. “New algorithm based on the hough transform for the analysis of pulsed thermographic sequences”. “Through skin sensing assessment of aircraft structures using pulsed thermography”. [67] D. 128 .V. Allport. S. A. 37. [60] X. [61] X.V. Composites Sci. A. Thompson and D. [63] P. 5. Sharp. “Quantitative evaluationof transient video thermography”.2698 (1996). Maldague. 79–123 (1998).P. Gonzalez. 175–81 (2002). Inverse Problems in Engineering. De Giovanni. Technol.C. 857–877 (2005). pp. A.P. Teruzzi. “Méthodes intégrales et contrôle non destructif par thermographie infrarouge stimulée“. 37. Didierjean . J. Bendada. A.A.P. [64] J. Infrared Phys. “Thermographic NDE of industrial materials and structures”. X. Ziadi. Therm. R. pp. [54] N. pp. 452-460 (1987). Almond.V. Taylor. Review of progress in quantitative nondestructive evaluation. C. [57] A. pp. R. Maldague. Maldague. noise. 12. D. QIRT Journal 4. Eval. Cielo. Rev Gen Therm 35. eds. Maldague. Couturier. 39. X. Jaeger. Maillet. F. “Double pulse infrared thermography”. D. Bendada. C. J. 353–362 (2004). Galmiche. J Infrared Phys. Ibarra-Castanedo. eds. pp. Oxford University (1959). Phys. Technol. A. “Conduction of Heat in Solids”.P. 559–564 (2004).[53] N. Y. Mater.S. 2nd ed.V. Deem. 297–301 (2002). Vavilov. Maldague.S. McHugh. [65] V. 137–53 (1993).P.S. [62] X. 43. NDT&E International 37. Marinetti. pp. M. J. Houlbert. Maldague. pp. R. Avdelidis.V. Chimenti. Ibarra-Castanedo. Klein. 1. 704-717 (1998).

H.T. Wu. 5-8. Allen. track 1264. H. Romani. Merla.M. Kuo. Maldague. “Automatic Interpolated Differentiated Absolute Contrast Algorithm for the Analysis of Pulsed Thermographic Sequence”. “Quantifying the Relevance and Stage of Disease with the Tau image Technique”. Thomas. Proc. 89-100 (2005). A. Infrared Physics and Technology 48. Gén. [76] D. 7th Conference on Quantitative InfraRed Thermography (QIRT). GESTS Int’l Trans.J Jin. X. Thermosense XV. G. 113-119 (1991).M. S. Klein. 343-347 (1996). 25. 11A.V. QIRT Journal 4. Proceedings CD. Hamann. Breitenstein.L. Romani. pp. pp. Levesque. Di Donato. [74] W. 71-76 (1987). López-Higuera. Di Luzio. Favro. O.P. Wong. Meola. G. “Evaluation of Defects in Composite Plates under Convective Environments using Lock-In Thermography”. L. Bendada. “Differentiated absolute phase contrast algorithm for the analysis of pulsed thermographic sequences”. 1. M. pp. Proceedings SPIE. pp. pp. 37.V.16. Meas. X. pp.E.Wang. L Di Donato. QUIRT-96. “Lock-in Thermography for NonDestructive Evaluation of Materials”. 13. J. Salerno. Madruga.6 (2004). [77] R.V.R. [70] D. Sci. 86-91 (2002). 85-105 (2007). T.P. Busse. [73] D. Photonics Spectra. C. Busse. Sci. “Thermographie IR: quelle longueur d’onde choisir I“.. 6. González. X. Deem. 1583-1590 (2002).X. 22. Y. Rhode Saint Genèse (Belgium). May. of the IEEE Engineering in Medicine and Biology 27th Annual Conference. 1. Ahmed. ed. [79] D. Wang. “Thermographie IR: quelle longueur d’onde choisir II“.P.J. “IR Thermal wave tomographic studies of structural composites”.A. “Unsteady-state lock-in thermography – Applications to shunts in solar cells”. “Characterization of electromagnetic fields using a lock-in infrared thermographic system”. G. Merla. B. P. 129 . [80] D. Balageas. Wu. 12. Maldague. [71] C.G. pp. Ibarra-Castanedo. Technol. Romani. Pilla. Computer Science and Engr. Pajani. pp. Chimenti. Carlomagno. Rev. Measures.1-H.A. 693-703 (1998). Ibarra-Castanedo.[69] D. F. Proc. Gonzalez. Shanghai (China) (2005). [75] C. of the 23th Annual International Conference of IEEE Engineering in Medicine and Biology. Ibarra-Castanedo . G.D. “Biomedical Applications of Functional Infrared Imaging”. Therm. H. 142-150 (2001). G. pp.16. 274-285 (1993). X.. G. [78] D. “Assistance to surgery by means of infrared functional imaging: preliminary results”. Gupta. Woolaway. Meas. 2. Proc. “Image and signal processing techniques in pulsed thermography”. 16–21 (2006).S. [82] A. “Nondestructive Control of Industrial Materials by Means of Lock-In Thermography”. Technol. [81] J. P. Istanbul (Turkey) (2001). A. “Lock-in thermography for imaging of modulated flow in blood vessels”. Bai. C. Review of progress in quantitative nondestructive evaluation. June. A. Squillace. A. [83] A. Pajani. IEEE Engineering in Medicine and Biology Magazine 21. 447-452 (1998). Proc. Measures. [84] A Merla. M. R. pp. Eurotherm Seminar 50. Maldague. [72] L. pp. L.L. 77-80 (1987). Giorleo.K. “New sensor technology for 3 to 5 μm imaging band”. pp. Thompson and D.

pp. Pavlidis. Romani. 73-79 (2002) [94] A. “Imaging the Cardiovascular Pulse”. L Di Donato. “Assessment of the effects of the varicocelectomy on the thermoregulatory control of the scrotum”. IEEE Engineering in Medicine and Biology Magazine. Sun. [91] A. Romani. P. G. Proc. 199-200 (2002). Merla. Pavlidis. [96] M. Fertility and Sterility 78. L Di Donato. G. G. Farina. 1418-1426 (2007). of the 23th Annual International Conference of the IEEE EMBS. Merla. of the 25th Annual International Conference of the IEEE EMBS. Fei. International Journal of Immunopathology and Pharmacology 15. Proc. [95] O. T. Arch Phys. Merla. Video. Cancun (Mexico) (2003). [90] A Merla. 416-421 (2005). Hilton Head Island. M. Garbey. P. Kiener. 2. Sun. V. Di Donato. Vacariu. 1. M. 2. L. Proc. A. P. 356-363 (2004). A. Pavlidis. pp. A. In International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI’06). pp. G. “Use of Infrared Functional Imaging to detect impaired thermoregulatory control in men with asymptomatic varicocele”. 569 -576 (2006). “Harvesting the thermal cardiac pulse signal”. Salsano. South Carolina (USA) (2000). [98] N. . F. G. pp. Pisarri. Garbey. Di Donato. G. J. Med. pp. pp. I. 1080-1081 (1999). I. of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’05) . Proceedings of the 2nd European Medical and Biological Engineering Conference EMBEC02. Ledda. H. 1. [89] A. Ledda. Rehabil. Fertility and Sterility. L.Rossini. 81 (2000). Proietti. “Contact-Free Measurement of Cardiac Pulse Based on the Analysis of Thermal Imagery”. “Study of Raynaud’s Phenomenon by means of infrared functional imaging”. [87] M. G. “Infrared Functional Imaging Evaluation of the Sympathetic Thermal Response”. 2. [86] I. Romani. L. G. Catalog No: 03CH37439C. 8. Merla. pp. Merla. Di Donato. Medical & Biological Engineering & Computing. Baukol. IEEE Workshop on computer Vision Beyond the Visible spectrum: Methods and Applications. [97] N. L. Merla. “Estimation of Blood Flow Speed and Vessel Location from Thermal”. 54. “Recording of the Sympathetic Thermal Response by means of Infrared Functional Imaging”. pp. [93] A. 104-109. [92] A. G. M. “Infrared Functional Imaging Applied to Raynaud’s Phenomenon”. S. Schuhfried. 41-52 (2002).[85] A Merla. A. M. 471-472 (2004). Romani et al. pp. A. I. Romani. 37. 81. Pavlidis. Fialka-Moser. Garbey A. F. Istanbul (Turkey) (2001). Merla. “Raynaud’s Phenomenon: infrared functional imaging applied to diagnosis and drugs effects”. of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’04). 6.. J. “Dynamic Digital Telethermography: A novel approach to the diagnosis of varicocele”. IEEE Engineering in Medicine and Biology Magazine 21. 1. N. [88] A. Rossini. Lang. Di Donato. 2. “Thermal imaging for anxiety detection”. “Thermographic Parameters in the Diagnosis of SecondaryRaynaud’s Phenomenon”. Salsano. Romani et al. Salsano. 1610-1611 (2002). F. 130 . Sun. G. Korpan. I. Levine. Pavlidis. Merla. A Ledda. Di Donato. Merla. pp. Garbey. Romani. Proc. P.. Romani.

van Assendelft. “ Psychological disabilities amongst patients with port wine stains”. [105] A. [114] W. 121. [113] L. Lamine. L. 2.W. Y. Buse. 3. pp. Milescu. The J. Yang. pp. 19. Na . Mater. G. Moyne.58-62 (2000). 209-15 (1989). [111] L. Bush A. Identification by measurement of thermal contrast”. A. van Gemert. pp. C.pp. “Perturbations Methods in Heat Transfer”. S.[99] M. N. Babalola.312-323 (2000). 93-104. Batsale. W.S. T. [103] S. Yassa. pp. Plenum Press.W. “Remarks on algorithm 368.W. Brown. pp. Degiovanni. ed. Stehfest. “ Vital sign estimation from passive thermal video”. 1110-1118 (1988). 69. Didierjean . “Approximate solutions to three-dimensional unsteady heat conduction through plane flaws within anisotropic media using a perturbation method”. T. 1-8 (2008). Dermatol. Eng. (1999). (1970). Yu. Invest. Milescu.G.J.30-42. “Temperature Profiles with Respect to Inhomogeneity and Geometry of the Human Body”. J. “The potential of dynamic area telethermometry in assessing breast cancer”. Bendada. Turner. C.A. 137–53 (1993). Anbar. pp. Andre.F. M. IEEE Conference. Br. A review”. Buursma. NY (1995). Belliveau. L. Werner. IEEE Eng. Med. 333-338 (1977). “Visible and Near Infrared Absorption Spectra of Human and Animal Haemoglobin”. [104] A. numerical inversion of Laplace transforms”. [100] J. Brown. Biol.J. IEEE Eng. Biol. [101] M. Welch. “Non-destructive thermal evaluation of delaminations in a laminate: I. 3698. A. VSP Publishing. Composites Sci. J. A. Houlbert. Myron. Dermatol. Chichester: Wiley (2000) [108] H. M. Aziz . Berlin: Springer (1984). P. Wu. 3. Utrecht (2000).C. Proc. Cohen. Gentner. E. Liu. Med. Sci. Keyerserlingk. Maillet. R. Computer Vision and Pattern Recognition. Lucassen. Douven. pp. SPIE. “Optical-Thermal Response of Laser-Irradiated Tissue”. FL:CRC (1992). L. [102] M. M.A. Babalola. 19. Degiovanni. 3. “Thermal Quadrupoles: Solving the Heat Equation Through Integral Transforms”. A. 673–84 (2002). [106] A. 624. . ”Perturbation Methods for Engineers and Scientists”. [112] A. Q. (2000). Lanigan. ”Clinical Applications of DAT using a QWIP FPA camera”. Ahlgren. 3914. P. ACM. SPIE Proc. 47. [107] D. D. “ Retrieval of opticalproperties of skin from measurement and modelingthe diffuse reflectance“. pp. J. J. Technol. Maillet.C. 10. [110] J. C. “Measureament of the thermal properties of the human skin. O. 131 . Journal of Applied Physiology 65. [109] D. pp. pp. Zijlstra. Commun. Modelling Simul. Boca Raton. A. Anbar. “Functional Infrared Imaging of the breast”. Cotterill. J. S.

C. Pickeringt. van Gemert. M. pp. 1247-1258 (1989). A. pp. “Computed temperature distributions around ectatic capillaries exposed to yellow (578 nm) laser light”.[115] A. Ring.P. 9. W. Biomed. Roggan. Biol. Phys. 4.J.36-46. Butler.J. 345-357 (1996) 132 . E. B. P. “Light Distributions in a Port Wine Stain Model Containing Multiple Cylindrical and Curved Blood Vessels”. Lucassen. 34. M. [117] G. Opt. Med. Lasers in Surgery and Medicine 18. Keijzer.V. [116] J.H. Walker. J. Hahn.W.. (1999). Muller. G. M. Friebl. “Optical properties of circulating human blood in the wavelength range 400-2500 nm“. W.

Sign up to vote on this title
UsefulNot useful

Master Your Semester with Scribd & The New York Times

Special offer for students: Only $4.99/month.

Master Your Semester with a Special Offer from Scribd & The New York Times

Cancel anytime.