See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/268821995

Simple method to measure gamma camera energy resolution

Article  in  Radiography · May 2015

DOI: 10.1016/j.radi.2014.10.007

CITATION READS

1 2,715

2 authors:

Nasser Ballani Geoff Dougherty

Lebanese University California State University, Channel Islands
33 PUBLICATIONS   151 CITATIONS    161 PUBLICATIONS   2,781 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Design and implementation of organ dose calculation in CT scanning View project

Digital Image Processing and Analysis of Complex Materials using Micro-CT View project

All content following this page was uploaded by Geoff Dougherty on 02 February 2018.

The user has requested enhancement of the downloaded file.

 Radiography 21 (2015) 172e175

Contents lists available at ScienceDirect

Radiography
journal homepage: www.elsevier.com/locate/radi

Simple method to measure gamma camera energy resolution

Nasser Ballani a, Geoff Dougherty b, *
a
Radiologic Sciences, Faculty of Public Health, Lebanese University, Beirut, Lebanon
b
Applied Physics and Medical Imaging, California State University Channel Islands, CA, USA

a r t i c l e i n f o a b s t r a c t

Article history: Purpose: Energy resolution is one of the major limitations of gamma camera performance, mainly
Received 2 August 2014 affecting image contrast and resolution. There is a need for a simple method of measuring gamma
Received in revised form camera energy resolution, which is practical for technology students as well as for routine quality
29 September 2014
control.
Accepted 17 October 2014
Available online 2 November 2014
Materials and methods: A 37 MBq (1 mCi) MBq 99Tcm point source was prepared and positioned 1.5 m
away from the gamma camera. Eleven static images were acquired with the same acquisition time (60 s),
using 1 keV windows at intervals of 5 keV from 115 to 165 keV. The counts for each image were recorded
Keywords:
Energy resolution
and plotted graphically (counts vs. energy). Gaussian fitting was used to estimate the full-width-at-half-
Imaging physics laboratory maximum height (FWHM), and the energy resolution (FWHM%) was calculated as a percentage of the
Dual-head gamma camera photopeak (~140 keV).
Image quality Results: The FWHM% of the energy peak was measured to be ~7%. Most values for the energy resolution
Gaussian fitting (FWHM%) of our system were significantly lower (i.e., higher energy resolution) than the commissioning
measurement, and were comparable to the recent preventive maintenance values. There were no sig-
nificant energy resolution differences between the two detectors of the dual head gamma camera.
Conclusions: This simple method for the evaluation of the energy resolution of a gamma camera system
can be easily implemented within routine quality assurance. The measure will help prevent image
deterioration through early detection of serious energy resolution issues which can be resolved in
routine gamma camera corrective maintenance.
© 2014 The College of Radiographers. Published by Elsevier Ltd. All rights reserved.

Introduction computed tomography (SPECT) cameras are reported in NU-1

The performance characteristics of a gamma camera should be
assessed at scheduled time intervals, and should include such
characteristics as energy resolution, intrinsic resolution, spatial
linearity, uniformity, and counting rate performance. Limitations in
the performance of the detector and its associated circuitry, and
with the collimator, are the main causes affecting image contrast
and resolution.1e3
The assessment of gamma camera performance characteristics
is standardized by the National Electrical Manufacturers Associa-
tion (NEMA). NEMA publishes documents that describe how to
perform, analyze, and report gamma camera performance charac-
teristics. Most recent NEMA standards for single-photon emission
* Corresponding author. Applied Physics and Medical Imaging, California State
University Channel Islands, Camarillo, CA 93012, USA. Tel.: þ1 8054378990; fax: þ1
8054372765.
E-mail address: geoff.dougherty@csuci.edu (G. Dougherty).
2007.4e6 In Europe a comprehensive description of procedures to
be performed in clinical departments has been developed by the
Institute of Physics and Engineering in Medicine (IPEM).7 Despite
becoming a standard reference in Nuclear Medicine Departments,
providing scientific staff with practical advice on the imple-
mentation of an effective quality control program, original sections
of the previous IPEM were revised and enhanced substantially by
report 867 to incorporate up-to-date developments in gamma
camera technology.
The NEMA recommendations facilitate unambiguous measures
of gamma camera performance,8 so that different systems can be
compared consistently. By contrast, other performance character-
istics rely on the manufacturer's initial specifications without any
guarantee of NEMA conformity.2,8 Energy resolution is expressed
by the spectrum broadening on the detector caused primarily by
random statistical variation of the events that form the output
signal.1 The width of the photopeak, DE, measured across its points
of half-maximum amplitude is the energy resolution; usually it is
expressed as a percentage of the photopeak energy, Eg:

http://dx.doi.org/10.1016/j.radi.2014.10.007
1078-8174/© 2014 The College of Radiographers. Published by Elsevier Ltd. All rights reserved.

Downloaded from ClinicalKey.com at Deaconess-Billings Clinic - JCon August 18, 2016.

For personal use only. No other uses without permission. Copyright ©2016. Elsevier Inc. All rights reserved.
 N. Ballani, G. Dougherty / Radiography 21 (2015) 172e175 173

FWHM (%) ¼ (DE/Eg)  100% (1)

NEMA recommendations for measuring energy resolution

require interfacing the gamma camera un-corrected signal (Z-
pulse) into an external multichannel analyzer or single channel
analyzer.2 Nowadays, energy resolution and correction software
which is provided by the manufacturer is usually camera-specific.
This often means that data from different cameras cannot be
conveniently measured and compared. AAMP recommends that
energy resolution is measured annually,9 and this is generally car-
ried out by the manufacturer as part of the warranty policy.
To avoid having to access the Z-pulse and potentially voiding the
camera's warranty, we have developed a straightforward method
for measuring and analyzing gamma energy resolution which
conforms to the most recent NEMA standards. A somewhat similar
method was suggested previously by Elliott,10 but neither practical
experimental work nor results were presented. As part of a physics
of medical imaging laboratory,11 our method has been used suc-
cessfully by nuclear medicine students for more than a decade
using Siemens E. Cam and GE Millennium gamma cameras. Our
method allows multiple detector system energy resolution data to
be analyzed in a consistent way, which is essential in comparing
performance.12,13 Energy resolution is known to vary with scintil-
lation crystal size, radionuclide energy, and field of view (FOV).14,15
Inter-camera energy resolution variability should be observed and
reported.
Materials and methods
Our preparation and acquisition settings adhered to the rec-
ommendations of the National Electrical Manufacturers Associa-
tion (NEMA), the Institute of Physics and Engineering in Medicine
(IPEM) and the American Association of Physicists in Medicine
(AAPM), i.e., the collimator was unmounted, the count rate was
within the system count rate capability (less than 15 K cps is a good
compromise), and a small source (less than 2 cm in diameter) was
placed at a distance of at least 2 crystal diameters along the central
axis. The outer 10% of the crystal area (or 5% of the diameter) was
masked with a circular lead mask, at least 3 mm thick, carefully
centered to ensure that the edge packing was covered. For 99Tcm,
the channel position of the center of the photopeak (140 keV) was
determined.
A 99Tcm point source of 37 MBq was prepared and positioned
1.5 m away from the crystal. Prior to running this experiment, the
field of view (FOV) intrinsic uniformity, which is routinely
measured at the same distance during quality control, was deter-
mined to be within the specifications. Eleven static images of 5 keV
Figure 1. Average counts acquired at one detector versus energy window (black), and
fitted to a Gaussian (red). (For 99Tcm point source).
intervals were acquired using a GE Millennium MP gamma camera
with the same acquisition time (60 s), using 1 keV windows at
intervals of 5 keV from 115 to 165 keV. Counts were repeated five
times at each energy window, and corrected for decay. The
counting uncertainty at each energy window was calculated.
Since energy resolution may vary at different energy levels, the
energy resolution was also measured for 131I (photopeak at
~364 keV).
Results
The counts acquired at each energy window and their averages
were tabulated. An example for one detector is summarized in
Table 1. ANOVA Statistical analysis shows no significant difference
between FWHMs of the five repeated sets of measurements a
FWHM grand mean of 10.03 and a pooled standard deviation of þ/
0.28. This confirmed FOV uniformity, and indicated that acquisi-
tion of a single set of measurements would be sufficient for further
work to obtain detector energy resolution.
The average counts at each window were plotted against the
corresponding energy, and Gaussian fitting was used to estimate
the full-width-at-half-maximum height, FWHM (Fig. 1). The energy
resolution (FWHM%) was calculated as a percentage of the photo-
peak (140 keV). A comparison between the energy characteristics of
two detectors within a dual head gamma camera is tabulated in
Table 2 and plotted in Fig. 2.
The counts vs. energy characteristic of a detector using a 131I
point source is tabulated in Table 3 and plotted in Fig. 3.

Table 1
Counts acquired at different energy windows using a99Tcm point source (The resulting FWHM and energy resolution, FWHM%, from Gaussian fitting is also tabulated).

Energy window Set#1 Counts/minute Set#2 Counts/minute Set#3 Counts/minute Set#4 Counts/minute Set#5 Counts/minute Average

115e116 122 110 125 119 121 119.4

120e121 100 112 105 111 129 111.4
125e126 175 180 171 182 178 177.2
130e131 253 276 281 259 249 263.6
135e136 1511 1570 1580 1525 1533 1543.8
140e141 3000 3058 3044 3005 2999 3021.2
145e146 2223 2222 2167 2228 2199 2207.8
150e151 755 721 766 747 750 747.8
155e156 123 125 125 112 126 122.2
160e161 15 17 14 10 12 13.6
165e166 20 22 18 15 14 17.8
FWHM (keV) 9.95 ± 0.26 9.99 ± 0.25 10.10 ± 0.30 10.05 ± 0.28 10.04 ± 0.30 10.04 ± 0.28
XCenter (keV) 141.09 ± 0.11 140.97 ± 0.11 140.93 ± 0.13 141.07 ± 0.12 141.03 ± 0.13 141.02 ± 0.12
Energy Resolution FWHM% 7.1% ± 0.2% 7.1% ± 0.2% 7.2% ± 0.2% 7.1% ± 0.2% 7.1% ± 0.2% 7.1% ± 0.2%

Downloaded from ClinicalKey.com at Deaconess-Billings Clinic - JCon August 18, 2016.

For personal use only. No other uses without permission. Copyright ©2016. Elsevier Inc. All rights reserved.
174 N. Ballani, G. Dougherty / Radiography 21 (2015) 172e175

Table 2 Table 3
Counts acquired from the two detectors of a dual-head imaging system at different Counts acquired at different energy windows using a 131I point source.
energy windows using a99Tcm point source. (The resulting FWHM and energy res- (The resulting FWHM and energy resolution, FWHM%, from Gaussian
olution, FWHM%, from Gaussian fitting is also tabulated). fitting is also tabulated).

Energy window Detector #1 Detector #2 Energy window Detector #1

115e116 122 222 340e341 37

120e121 100 255 345e346 62
125e126 175 235 350e351 85
130e131 253 496 355e356 112
135e136 1511 1434 360e361 150
140e141 3000 3125 365e366 172
145e146 2223 2248 370e371 200
150e151 755 750 375e376 192
155e156 123 15 380e381 150
160e161 15 3 385e386 123
165e166 20 3 390e391 111
FWHM (keV) 9.95 ± 0.26 9.69 ± 0.53 395e396 53
XCenter (keV) 141.09 ± 0.11 141.05 ± 0.23 FWHM (keV) 29.04 ± 3.41
Energy Resolution FWHM% 7.1% ± 0.2% 6.9% ± 0.3% XCenter (keV) 370.85 ± 0.56
Energy Resolution FWHM% 7.8% ± 0.9%

Discussion
The ability of the scintillation camera to distinguish between
photons of different energies is best expressed by energy resolu-
tion. It is conventionally quantified as the percentage ratio of the
full width at half-maximum height, FWHM, of the primary radia-
tion to the photopeak energy, FWHM%.3 In normal operation the
photopeak profile and its peak energy are not readily accessed by
the operator. In this work, we have demonstrated how they may be
obtained simply.
In the camera system it is possible to discriminate against
scattered photons using pulse height analysis. The window setting
of the pulse height analyzer (PHA) determines the amount of
scattering registered in the image; a narrower window setting
dramatically eliminates scattered photons, and consequently im-
proves image quality in terms of contrast and (spatial) resolution.1
The energy resolution of the detector is the most important feature
that determines the efficiency with which this can be accom-
plished. With a smaller detector energy resolution (FWHM%), it is
possible to set a narrower window setting and achieve better image
quality. For example, with a FWHM% of ~10% the window can be set
to 130e150 keV (~15%), which is a common setting for clinical
studies.
Many factors degrade the gamma camera energy resolution,
such as poor matching of the gains of the photomultipliers,
photomultiplier(s) failure or malfunction, crystal defects, physical
separation of the photomultiplier-light guide assembly from the
crystal, and high count rate.3 Changes in gamma camera sensitivity
could indicate incorrect energy calibration of the PHA or impaired
energy resolution.3 An immediate evaluation of the gamma camera
energy resolution is recommended in such an eventuality to
determine which is the case. Such an evaluation can be confirmed
in an effortless and expeditious way.
In order to improve the overall energy resolution, gamma
cameras have electronic methods of energy correction that align
the photopeaks produced by each of the photomultiplier tubes. This
is usually done by energy correction tables which are periodically
(annually) recollected over the life of the camera, ensuring opera-
tional improvement.3 Our straightforward method of determining
the energy resolution can indicate whether a manufacturer service
or update of the proper energy correction is warranted.
Energy resolution is dependent on detector crystal thickness.
Better energy resolution can be obtained at thin crystal size, (3/8”
(9.5 mm) compared to 5/8” (15.8 mm)). Energy resolution is more
variable among different gamma cameras using a thick crystal.
However, energy resolution should not be variable in multiple
detectors within the same imaging system.11 In this experiment we
found the energy resolution of all our departmental gamma cam-
eras using 99Tcm to be similar within NEMA specifications (FWHM
% < 10%).

Figure 2. Gaussian fitting of the counts acquired at the two detectors (detector #1 in
black; detector #2 in red) of a dual-head imaging system versus energy window. (For Figure 3. Counts acquired at one detector versus energy window (black), and fitted to
99 m
Tc point source). a Gaussian (red). (For 131I point source).

Downloaded from ClinicalKey.com at Deaconess-Billings Clinic - JCon August 18, 2016.

For personal use only. No other uses without permission. Copyright ©2016. Elsevier Inc. All rights reserved.
 N. Ballani, G. Dougherty / Radiography 21 (2015) 172e175
Since the number of photoelectrons released from the photo-
cathode increases with increasing photon energy, energy resolu-
tion should become better with increasing gamma-ray energy.
However, most of the functions of a gamma camera are optimized
for 99Tcm. This explains why we obtained the same energy resolu-
tion for 131I (364 keV). (This was also within NEMA specifications).
The increased gain used at higher energy is probably the main
reason for the elevated center value obtained for 131I, viz. 370 keV
rather than the expected 364 keV, although the arbitrariness of the
positioning of the energy windows will also contribute to a dif-
ference between the center value obtained and the expected value.
The latter reason also holds for the 99Tcm results.
Although dual- or triple-head gamma cameras present more
quality control concerns, similar performance characteristics
among detectors within the same multi-head gamma camera
would be reassuring for qualitative and quantitative imaging pro-
cedures.10,11,16 For the dual-head gamma cameras tested here, the
energy resolution was the same for both detectors. The energy
resolution measurement method which had been suggested pre-
viously by Elliot10 did not result in any experimental data, so that
we were unable to compare the method with ours.
Continuous gamma camera quality control tests are essential.
Proper testing can determine whether performance is acceptable.
Measurement of energy resolution at commissioning, and during
preventive maintenance and essential quality control times must
be performed. This measurement is a task for the medical engineer
and/or medical physicist, although a professional technologist
should also be familiar with the methodology and take turns on
such a task. In certain countries the status of professional tech-
nologist would require energy resolution measurement to be part
of his/her role.11
Conclusion
Experimentally, we have demonstrated a simple method and
procedures for measuring the energy resolution of a gamma cam-
era. It gives technologists, both students and practitioners, the
opportunity to optimize the gamma camera detector for clinical use
before imaging human subjects. Using this method, the energy
Downloaded from ClinicalKey.com at Deaconess-Billings Clinic - JCon August 18, 2016.
View publication stats For personal use only. No other uses without permission. Copyright ©2016. Elsevier Inc. All rights reserved.
175
resolutions of the gamma cameras in the department were
confirmed to be  10% FWHM.
Conflict of interest
None.
References
1. Simon RC, Sorenson JA, Phelps ME. The gamma camera: performance charac-
teristics [chapter 14]. In: Physics in nuclear medicine. Philadelphia: Elsevier/
Saunders; 2012.
2. IAEA-TECDOC-602. Quality control of nuclear medicine instruments 1991.
Vienna: International Atomic Energy Agency; 1991.
3. IAEA-TECDOC-xxxx. Quality control of nuclear medicine instruments 2001.
Vienna: International Atomic Energy Agency; 2000.
4. NEMA Standards Publication NU1-2007. Performance measurements of gamma
cameras. Rosslyn, VA: National Electrical Manufacturers Association; 2007.
5. Hines H, Kayayan R, Colsher J, Hashimoto D, Schubert R, Fernando J, et al.
Recommendations for implementing SPECT instrumentation quality control.
Nuclear medicine SectioneNational Electrical Manufacturers Association
(NEMA). Eur J Nucl Med 1999 May;26(5):527e32.
6. Groch M, Erwin WD. Single-photon emission computed tomography in the
year 2001: instrumentation and quality control. J Nucl Med Technol 2001;29:
9e15.
7. Bolster A, editor. IPEM Report 86, Quality control of gamma camera systems.
York: Institute of Physics and Engineering in Medicine; 2003.
8. Rova A, Celler A, Hamarneh G. Development of NEMA-based software for
gamma camera quality control. J Digit Imaging 2008;21(2):243e55.
9. AAPM Report No. 6, Scintillation camera acceptance testing and performance
evaluation. New York: The American Association of Physicists in Medicine;
1980.
10. Elliott AT. Quality assurance. In: Sharp PF, Gemmell HG, Murray AD, editors.
Practical nuclear medicine. Springer; 2005. p. 78 [chapter 6].
11. Ballani NS, Sukkar I. Medical imaging physics teaching to radiologic technol-
ogists in Kuwait. Radiography 2005;11(1):67e70.
12. Grosev D, Loncari c S, Vandenberghe S, Dodig D. Triple-head gamma camera
PET: system overview and performance characteristics. Nucl Med Commun
2002 Aug;23(8):809e14.
13. American Society of Nuclear Cardiology. Updated imaging guidelines for nu-
clear cardiology procedures, part 1. J Nucl Cardiol 2001;8(1):G5e58.
14. Kappadath SC, Erwin WD, Wendt 3rd RE. Observed inter-camera variability of
clinically relevant performance characteristics for Siemens Symbia gamma
cameras. J Appl Clin Med Phys 2006;7(4):74e80.
15. Bugby SL, Lees JL, Bhatia BS, Perkins AC. Characterisation of a high resolution
small field of view portable gamma camera. Phys Medica 2014;30:331e9.
16. Blagosklonov O, Sabbah A, Verdenet J, Baud M, Cardot JC. Poststress motionlike
artifacts caused by the use of a dual-head gamma camera for (201)Tl
myocardial SPECT. J Nucl Med 2002;43(3):285e91.