You are on page 1of 6

ITCH - ELECTRO 2007 Octubre 17-19, Chihuahua, México


Víctor Hugo Ortiz Flores2, Cornelio Yánez Márquez1, Jorge Isaac Chairez Oria2
1Centro de Investigación en Computación.
Ave. Juan de Dios Batíz, s/n Colonia Nueva Industrial Vallejo
C.P. 07738 México, D.F.,
2 Unidad Profesional Interdisciplinaria de Biotecnología
Ave. Acueducto de Guadalupe, s/n Colonia Barrio la Laguna Ticoman
C.P. 07340 México, D.F.

ABSTRACT It does not require specific treatment, although it

In this paper an algorithm is developed that allows must alert to the clinical one with respect to the
by means of the method of finite element to detect intensity of the skull traumatism.
skull fractures. We intend the theoretical
foundation that sustains when segmenting cranial Fracture with collapse. Here, there is a depression
images, these methods based on the Method of of a bone fragment of the thickness of the skull.
Finite Element (FEM), for their initials in English
that allows us to detect fractures in the images. Simple or closed. This appears when the hairy
With the development of this paper the efficiency leather that covers the fracture remains intact.
of the Method of Finite Element is shown in the
segmentation of images. Likewise, it is made Composed or opened, when the hairy leather is
notice the importance that up to now have lacerate. It represents 80% of the fractures with
acquired in the areas of segmentation of images collapse. According to its cause and aspect, it
due to their capacity to solve numeric methods. In could be sub-classified in: armor-piercing,
turn, continuity is given to that the investigation penetrating associated to linear fractures or
groups that develop projects related with the comminutes. Also, they can be associated to
analysis of images prescribe, continue to the so lacerations of duramatter that constitutes a front
much of the area of finite element and they follow door for the infection. Mostly of cases require
them incorporating in their algorithms, mainly in debrillation and surgical elevation.
concerning tasks to the theory analysis of images
and their applications.. The progressive fractures or, badly called,
leptomeníngeos cysts (better
INTRODUCTION. pseudomeningoceles), take place by a defect in
Usually, strong blows in the head provoke duramatter under a cranial fracture, with the
changes on the skull structure at the impact point. consequent risk of exit of cephalon-raquideum
In several occasions, small objects can penetrate liquid and, sometimes, cerebral weave. Its
the skull and produce a local laceration of incidence is low, around 0.6% to 1% of the
encephalon. If big objects are applied with great fractures, being more frequent while the patient is
force can make some pieces or bone’s fragments younger. Habitually, they require surgical
be introduced themselves in encephalon in the site attention.
of impact. These reasons lead to develop an
adequate method to realize an opportune detection Fractures of frontal bone. - They take place by a
[1]. severe traumatism on the frontal region of skull.
The frontal sine can be jeopardized, and if the
1.1. Fracture types. later wall of the sine is fractured, injury of
duramatter and of the nasofrontal conduit can also
In general, the fractures can be classified in: exist [2].

Linear fracture. This is an elastic deformation of

the skull. It represents 80% of the skull fractures.

ITCH - ELECTRO 2007 Octubre 17-19, Chihuahua, México

1.2. Segmentation of Images. characterizing the physical behavior for a selected

problem in a series of non-intersecting sub-
The segmentation of images can be defined as the dominions to each other, denominated finite
process to subdivide or divide an image in elements. The set of finite elements forms a
regions, components, parts or objects. The partition of the dominion also denominated
segmentation can also be defined as finding, by discretization [4]. Within each element, a series of
means of a numerical algorithm its homogenous representative points called nodes is distinguished.
regions and their edges. Some algorithms for the Two nodes are adjacent if they belong to itself
labeled one of connected components will also be finite element, in addition a node on the border of
analyzed. Whatever it is the definition that on a specific finite element can belong to several
segmentation is adopted, the basic idea consists of elements. The set of nodes considering their
isolating in the image the different objects to be relations of adjacencies is called mesh. The
recognized. It is clear that the result of this calculations are made on a mesh or discretization
process can determine the performance of the created from the dominion generating some
complete system to analyze images. A bad special meshes distributions (this is done in a
segmentation in general will cause a bad previous stage to the calculations that denominate
recognition of objects inside the image. A good pre-process). In agreement with these relations of
segmentation, on the other hand, will cause that adjacencies or connectivity, the value of a set of
the SRAO gives good results. A formal way to unknown variables defined in each node is related
define the segmentation process is the following and denominated degrees of freedom. The set of
one [3]: The segmentation is the process of divide relations between the values of variable
an image f ( x, y ) in regions of pixels, so that determining between the nodes can be written in
form of system of equations linear (or linearized),
each one of these sub-images represents an object the matrix of this system of equations is called
or a part of itself. The segmentation is, then, the matrix of elasticity of the system. The number of
process that consists of grouping pixels in regions, equations of this system is proportional to the
R1 , R2 ,K , Rn such that: number of nodes [5].

a) U Ri ⊆ f ( x, y ) The mathematical description of the procedure to
be followed can be summarized by: The vectors of
i =1
image characteristics used in this work are
b) R1 , i = 1, 2,K n is connected profiles of derivative functions. They are vectors
c) Ri I Rj = ∅, ∀i, j , i ≠ j. whose components are directional derived which
are calculated in points of the image where those
d) P ( Ri ) = TRUE , i = 1, 2,K , n. Each points are on a straight line segment. In order to
Ri satisfies a predicate with some set reduce the effect of the noise in the images, it is
advisable to apply them a filter. Thus, E
properties. All the elements of each Ri represents the function of intensities of the image
share he himself set of properties. and Gσ is a Gaussian filter. The directional
e) P ( Ri U R j ) = FALSE for i ≠ j . This derived from the filtered image Gσ * E in the
means that pixels belonging to given
adjacent regions and being as a single point ( x, y , z ) , in the direction of the unitary
one does not satisfy the predicate. If it is vector n , is given by:
not then they will be considered like a
single region.
Dn ( Gσ * f )( x, y, z ) = n ' ∇ ( Gσ * f )( x, y, z ) (1)
1.3. Method of the Finite Element.
The Gaussian filter is described like:
This method is based on dividing the body,

(i 2
+ j2 )
structures or dominion (average continuous) on Gσ ( i, j ) = c e 2σ 2
which are defined certain integral equations

ITCH - ELECTRO 2007 Octubre 17-19, Chihuahua, México

Where: We propose the Energetic Function.

E = ∑  pi , j ,k − xi , j , k 
σ = standard deviation. The constant c one
i, j
calculates so that the coefficients add 1. The 144 42444 3
greater σ , the smoothness property is increased. data fidelity term

 2
In the vertex vi of a mesh, the normal unitary (
+ λ  ∑ ∑ xi , j ,k − xis , js ,ks  )
 i , j ,k is , js ,ks 4244444
14444 3
outside calculates ni to the mesh and other
enforces smoothness (3)
unitary vectors are selected conforming a same
angle with ni . This creates and uniformly Where:
distribution and it is distributed around ni , in pi , j ,k is a voxel value from the original image.
order to obtain a set N v of directions. xi , j ,k is the classification of that voxel.
is js ks represent the neighborhood of the voxel.
Secondly the image is segmented. In this case we
chose the method of canny. The operator of edges The energy is a numerical value representing a
of Canny first is derived from the Gaussian filter. weighted summation of two main distances to the
This operator approximates the operator strongly,
voxels taking part in the segmentation function of
optimizing the product of the quotients signal to
intensities [8]. In this study, it is desirable to make
noise and location [6].
a detailed registry for all images and considering
the derivates form each image structure. This can
Thirdly we began the Method of the Finite be solved by the definition of vector field that can
Element. Taken to a closed enclosure the steps for
vary point to point in the images. In this case, the
the resolution they are [7]:
energetic function can be redefined as
E = ∑  pi , j ,k − xi , j , k 
• To divide the enclosure in Finite
Elements: Triangles (3 nodes), i, j
Tetrahedrons (4 nodes), etc.
 2
• To deduce the equation that describes the
potential f within an EF.
+ λ  ∑ ∑ xi , j ,k − xis , js ,ks  )
 i , j ,k is , js ,ks 
• To raise the equations that gives the
conditions of adjustment of the solutions  2 
in the borders of the EF.  ∫ ∇u ( p ) d Ω 
• To calculate the potentials in the nodes of Ω 
each EF by means of some of the  2 
methods that will be in traduced below.
+  λx λ y λz   ∫ ∇v ( p ) d Ω 
• To solve the raised algebraic equations. Ω 
 ∇w ( p ) 2 d Ω 
 Ω∫
Generation of the finite Elements
• The contours can be irregular 
• The EF will be as small as the And where the updated coordinates in the
programmer considers. resulting image are:
• If the potential varies a lot, the EF will
consider a mesh with small “holes”. It
means the nodes are closer.

ITCH - ELECTRO 2007 Octubre 17-19, Chihuahua, México

 xˆ   x + u ( x, y, z )  RESULTS
The Figure 2, shows 3 types of image of people
 
E =  yˆ  =  y + v ( x, y, z )  who present anomalies and that can present
fractures that cannot be, so easily detected.
 zˆ   z + w ( x, y, z ) 
 x + ∑m N ( x , y , z ) ui , j . k 
 i =1, j =1 i , j , k

= y + ∑ i =1, j =1 N i , j ,k ( x, y, z ) ui , j.k 
 m

 
 
 z + ∑ i =1, j =1 N i , j ,k ( x, y, z ) ui , j ,k 

Figure 2. Three types of image
Where the interpolation or form functions
( N ( x, y , z ) )
i , j .k are traditionally used by the
In order to eliminate the noise acquired by the
acquisition, in addition to initialize with the
method of finite elements for rectangular meshes vertices of the finite element, the Gaussian filter
[14]. In general, m is the number of total nodes was applied with σ = 10 (Figure 3).
depending on the number of nodes forming each
element and the total number of elements which
defines the mesh. Since the dominions in the
images are generally of parallel sides, we have
used regular lagrangians elements of 8 nodes for 3
dimensions and 4 nodes for 2 dimensions. The
interpolation functions can be written easily,
based on the space coordinates of the image, by
simplicity we are going to define the system Figure 3. Images of Skull, with Gaussian filter.
coordinate in 2 dimensions within an element
anyone as it is in Figure 1 [9]. In order to observe some details of the curved
images, and edge canny detection was used.
Under this analysis, some relevant aspects for
each figure were observed. For example, lines,
among others (Figure 4).

Figure 1. - Definition of the system of


To avoid the separate computation of the forces F,

the elastic deformation, and the matching Figure 4. Images of Skull segmented.
criterion, we propose to directly compute a
deformation field that readily satisfies both Also, it was applied an alternative method to show
aspects, the elasticity constraint and a local image the vectors at the edge of the images (equation 1).
similarity constraint between the images to be These vectors can be helpful to design an
matched (I1 and I2). Hence, the total energy to be adequate mesh to initialize the Finite Element
minimized is expressed as: procedure [9]. (Figures 5 y 6)

E = ∫ λ t d Ω + ∫ I1 ( x + u ( x ) ) − I 2 ( x ) d Ω )

Ω Ω (6)

ITCH - ELECTRO 2007 Octubre 17-19, Chihuahua, México

Figure 5: Growing sphere. a) and b): close-ups of

2D cuts through 3D image with a) classical OF,
and b) FE matching deformation fields overlaid,
c) and 3D orthogonal cuts through the FE mesh
with intensity coding of the displacement field. Figure 8. Surface 2 of the Finite Element.
The displacement field is mainly located at the
boundaries of the sphere and is propagated Once the adequate mesh (by trial and error
through the surrounding elastic medium. method) has been applied to solve the finite
element, it is possible to observe small details in
the images that can be considered fractures
(confirmed by a-priory knowledge of where the
fracture is). This could be an interesting method to
help physicians without experience that are not so
experts to detect the fractures form the magnetic
resonance (Figure 9).

Figure 6: Enlarging ventricles. a) slice of

difference between segmented images at both time
points (gray means no difference), b) deformation
field superimposed on same image at the first time
point. c) close-up

In the following figures (7 and 8), it is showed the

surfaces of the finite element solution. These Figure 9. Images with surface 1 of the Finite
surfaces are composed by 8 nodes and 9 elements Element
of nodes, for the first and 16 nodes and 18
elements for the second one representing a With the second surface that has a large number
particular zone where a fracture is located. of nodes and therefore a “better” mesh, it is
. possible to appreciate similar fractures (Figure

Figure 10. Images with surface 1 of the Finite


Figure 7. Surface 1 of the Finite Element Here it is clear that a method to determine the
adequate amount of nodes and the elements form
must be designed. Some interesting approaches
can be used to realize this aspect: the neural
networks ideas, the genetic algorithms, etc

ITCH - ELECTRO 2007 Octubre 17-19, Chihuahua, México

CONCLUSIONS [6] Sossa J., Rasgos Descriptores para el

This paper presents a new, physics-based Reconocimiento de Objetos, Ciencias de la
deformable model for tracking physical Computación, IPN-UNAM-FCE, México, 2006
deformations (Fractures) using image matching.
The model results from the minimization of a [7] Passaro A., Júnior J., Abe N., Sasaki M.
deformation field simultaneously satisfying the and Santos J., Finite-Element and Genetic
constraints of an elastic body and a local image Algorithm Design of Multi-segmented Electro-
similarity measure. The model provides us with a optic Sensor for Pulsed High-Voltage
physically realistic deformation field and also Measurement, 6th World Congresses of
allows us to inspect the characteristics of the Structural and Multidisciplinary Optimization,
deformed objects. This can be very useful for the 2005.
inspection of stresses induced by the deformation
of certain objects on their surroundings, in our [8] Sanchez H. and Bribiesca E., Medida
case the detection of skull fractures. de Similitud para Objetos 2D y 3D a través de
una Energía de Trasformación Óptima,
In the experiments, the objects were considered to Computación y Sistemas, Mexico city, 2003, pp.
be homogeneous elastic bodies. Further 66-75.
improvements of the algorithm include the
assignment of different elasticity’s to the different [9] Yushkevich P., Piven J., Cody H., Ho S.
objects represented in the image. This will require and Gerig G., User-Guided Level Set
a preliminary segmentation of the objects to be Segmentation of Anatomical Structures with
deformed so as to be able to set appropriate ITK-SNAP, MICCAI Workshop, 2005.
elasticity coefficients to every cell of the mesh.
Also, the anisotropy of certain skull fractures [10] Reig S., Penedo M., Pascau J., Sánchez
could be included into the model by modifying the J., Arango C. and Desco M., Evaluación de dos
elasticity matrix D appropriately. métodos automáticos para la segmentación del
volumen intracraneal en imágenes de RM:
We would lack to design an algorithm that assures McStrip y BET, XXIV Congreso Anual de la
that the meshes are the best possible one (in the Sociedad Española de Ingeniería Biomédica,
sense of the image representation), also to include pp.495-498, 2006.
one better area within the image, providing an
improvement in the diagnostic supplied by the [11] Ortuño J., Malpica N., Reig S., Martínez
method suggested in this manuscript R., Desco M. and Santos A., Algoritmo
Morfológico de Segmentación de Imágenes de
BIBLIOGRAFY Resonancia Magnética Normales Ponderadas en
[1] Snell R., Neuroanatomía Clínica, T1, XIX Congreso Anual de la Sociedad Española
Medica Panamerica, 2001. Ingeniería Biomédica, pp.205 – 208, 2000.

[2] et.- , Harrison Principios de Medicina [12] Smith S., BET: Brain Extraction Tool,
Interna, Mc Graw Hill, 2006. Technical Report, UK, 2001.

[3] Gonzalez R. and Woods R., Digital [13] Lira J., Introducción al tratamiento
Image Processing, Prentice Hall, Inc., New digital de imágenes, Ciencia de la Computación,
Jersey. 2002. IPN-UNAM-FCE, México, 2002.

[4] Felippa C., Introduction to finite

element methods. Departament of Scientist,
Engineering y Aerospace Colorado University.
Boulder, 2001.

[5] Ledesma A., Fundamentos del Método

de Elemento Finito para Problemas Elípticos,
UNAM, 2006.