Добавил:
Sekretar
kiopkiopkiop18@yandex.ru
t.me/Prokururor I Вовсе не секретарь, но почту проверяю
Опубликованный материал нарушает ваши авторские права? Сообщите нам.
Вуз:
Предмет:
Файл:Ординатура / Хирургия / Библиотека им академика М.И. Перельмана / Книга_5545_Библиотеки_им_академика_М_И_Перельмана.pdf
124
Z XXYY=+++
+−+−
k
Z
XX=−
+−
k
Z
YY=−
+−
9 Gamma Camera
capacitance values are assigned in direct proportion to the location of the PM tube
relative to the four signals. Suppose a γ-ray interacts at a location (*) near tube 7.
The largest amount of light is received by tube 7, and other tubes receive light in
proportion to their distances from the point of interaction. The output signals of PM
tubes are weighted by the appropriate capacitance values and then summed to form
each of the X+, X−, Y+, and Y− signals individually. In this case, X− will be greater
than X+, and Y+ will be greater than Y−, because the interaction occurred in the upper
left quadrant. The X-, Y-designating pulses, X and Y, and the Z pulse are then
obtained as follows:
(9.1)
X
(9.2)
(9.3)
where k is a constant and k/Z is the amplier gain. The X and Y pulses are then projected on a display monitor to depict the X, Y coordinates of the point of γ-ray interaction, which in turn corresponds to the coordinates of the location in the eld of
view where the γ-ray interacted in the crystal. Similarly, these pulses can be stored
in the computer in a square matrix so that the data can be processed later to reproduce an image. Details of data acquisition and storage in the computer are given in
Chap. 11. Or, they can be projected on an x-ray lm. Nowadays, resistors and
microprocessors have been used in place of capacitors.
The larger the number of PM tubes, the better the accuracy of the X, Y locations
of pulses on the image; that is, the better the spatial resolution of the image (see
Chap. 10).
9.1.6 Pulse-Height Analyzer
After the Z pulses are formed by the summing circuit, the PHA analyzes their amplitude and selects only those of desired energy by the use of appropriate peak and
window settings. In many gamma cameras, the energy selection is made automatically by push-button-type isotope selectors designated for different radionuclides
such as
are selected by menu-driven algorithm on a computer monitor interfaced with the
camera. In some gamma cameras, two or three PHAs are used to select simultaneously two or three γ-rays of different energies. These types of cameras are useful in
imaging with
dow settings are expressed in percentages of the peak energy and for most studies,
a 15–20% window centered symmetrically on the photopeak is employed.
energy range selected by the PHA.If the Z pulse is outside this range, then X and Y
pulses are discarded.
99m
131
Tc,
I, and so on. In modern cameras, isotope peak and window settings
111
In and 67Ga that possess two or three predominant γ-rays. The win-
It should be noted that X and Y pulses are accepted if the Z pulse is within the
9.2 Digital Camera
125
9.1.7 Display andStorage
In a typical nuclear medicine study, data are collected normally for preset counts
(e.g., 500,000 counts) or a preset time (e.g., 10min). Until the mid-1990s, image
data were captured on x-ray lm or Polaroid lm, or stored on magnetic tapes, laser
disks, and the like. Nowadays, all camera systems use computer memories for storage of image data. The details of storage in computers are given in Chap. 11. In
older systems, images were mostly displayed on cathode ray tube (CRT) monitors
and at present, all systems commonly use LCD (liquid crystal display) video monitors for better display of images. The computer manipulation of image contrast on
LCD monitors provides a better view of images leading to more accurate diagnosis
of diseases. The details of display and storage are given in Chap. 11.
9.2 Digital Camera
It is seen from the above description that the X- and Y-pulses are obtained in analog
form and are projected on different display and recording systems. Such analog
processing inherently includes instability in pulse formation and results in image
nonlinearity and nonuniformity. These are caused by uctuations in PM tube output
due to high-voltage (HV) variations, drift in preamplier output, and variations in
PH and X-, Y-positioning analyses. To circumvent these effects and also for the
manipulation of data at a later time, analog data are digitized to be stored in a matrix
map in a computer. Digitization of the analog signal is performed by an electronic
circuit, called the analog-to-digital converter (ADC). The digitized data are later
retrieved for further processing to display on video monitors.
In modern cameras each PM tube output is digitized by the ADC before PH and
X-, Y-positioning analyses. These cameras are called “all-digital” cameras. In these
cameras, the gains of all PM tubes are initially optimized by placing a narrow beam
of a radioactive source in front of each PM tube and determining the center of the
photopeak by adjusting the high voltage of the PM tube with a digital computer.
Next, the camera is calibrated, in which a source of interest is positioned in front of
each PM tube, and output from each PM tube is sampled, integrated, and digitized
by a high-speed ADC in the computer. Each signal is then normalized by dividing it
with the sum of all digital signals arising from the same scintillation event. In a twodimensional array of PM tubes, the normalized digital output Zi(X, Y) corresponds
to the X, Y location of the PM tube i. To determine the location of each signal Zi, a
weighting factor is calculated from the inverse of the uncertainties of X and Y positions, that is, 1/ΔX and 1/ΔY, that are related to the spatial distribution of Zi values
around the center of the PM tube. The X, Y locations and weighting factors are
mapped and stored in reference tables as functions of Zi values for all PM tubes for
positional and Z-pulse analyses of a scintillation event in later imaging studies.
In subsequent patient imaging studies, the output signal of each PM tube from a
scintillation event is sampled, integrated, digitized and nally normalized to give Zi.
The location (X, Y) of the scintillation event is then calculated by using the appropriate values of locations and weighting factors in the reference tables in the memory.
126
9 Gamma Camera
The digitized Zi (X, Y) is stored in the X, Y location of the image matrix, if the pulse
discrimination does not reject the signal. Since the location of each event is determined by digitizing and analyzing the individual signal from each PM tube, the
accuracy of positioning of the signals is greatly improved. For these reasons, the
digital cameras provide excellent intrinsic linearity and hence superior spatial resolution in image formation.
9.2.1 Solid State Digital Camera
Digirad Corporation has made several commercially available several gamma cameras using solid state detectors. Initially CZT detectors were used, but later the company replaced them with CsI(Tl) detectors. Each detector head is pixelated with a
dimension of 21×16cm. There are 768 pixels in each head and each pixel (voxel)
has a dimension of 6.1×6.1mm. The uniqueness of these cameras is that they do
not use PM tubes for pulse formation but instead use silicon diodes. No X, Y positioning circuit is used, because each CsI(Tl)/silicon diode element functions as an
individual detection system, independent of other elements, and each event of photon interaction in the crystal is positioned in the image matrix corresponding to the
location of the element (Early 2005). This provides an excellent spatial resolution
and quality of the images in the energy range of 60–300keV.
Various Digirad camera models include Cardius X-ACT, Cardius 3 XPO (threehead), and Cardius 2 XPO (two-head). Appropriate collimators are required for
imaging different organs and photons of different energies. Many units are small
and portable. These cameras are commonly used for cardiac SPECT studies with
the use of a rotating chair for the patient. During the study, the patient is positioned
in the chair in front of the vertically standing camera and the chair rotates at incremental angles with respect to the detector providing desired projections.
Spectrum Dynamics has introduced a gamma camera (D-SPECT) using CZT
semiconductor as the detector for cardiac studies. It consists of an array of nine
columnar detectors that are arranged in a conguration to conform to the contour of
the left side of the patient’s chest. Each detector consists of 1024 (16×64) 5-mm
thick CZT crystals of size 2.46×2.46mm, and can rotate and translate individually
to obtain the desired number of angular projections around the patient. The data
acquisition is quite fast requiring only two minutes for a gated cardiac study and
providing high-quality images.
Another gamma camera using CZT crystals has been introduced by GE
Healthcare (Discovery NM 530c), primarily for cardiac studies. A focused multipinhole collimator is used to improve the detection efciency and hence sensitivity.
The detector and the collimator are held in a xed position so that many cardiac
projections are acquired simultaneously. This camera allows much faster acquisition of data (4–5min compared to 15–20min for conventional cameras), virtually
eliminating the artifacts caused by patient movement and also facilitating dose
9.3 Questions
127
reduction to the patient. Recently for fusion of anatomical and functional images, a
CT unit (LightSpeed VCT) has been incorporated in this system to make an integrated SPECT/CT unit, NM/CT 570c, to provide photon attenuation correction and
better delineation of lesions in organs.
9.3 Questions
1. (a) Describe the operational principles of a gamma camera.
(b) The main purpose of a collimator is to limit the eld of view of an imaging
device for imaging. True or false?
(c) The purpose of a photomultiplier tube is to convert light photons to an elec-
tron pulse. True or false?
(d) The scattered photons are excluded by the proper choice of a collimator.
True or false?
(e) Scattered photons are excluded by the proper choice of discriminator set-
tings (windows). True or false?
2. (a) What are the different categories of collimators?
(b) Which collimator is most used in nuclear medicine?
(c) Which types of collimator give image distortion and why?
3. Describe the function of the X, Y circuit in the gamma camera system.
4. A pulse-height analyzer:
(a) Reduces the background. True or false?
(b) Rejects γ-rays that undergo Compton scattering in the patient and the detec-
tor. True or false?
(c) Rejects γ-rays undergoing photoelectric effect in patients. True or false?
(d) Increases the signal-to-noise ratio. True or false?
5. (a) The detection efciency of a gamma camera increases with the thickness of
the detector. True or false?
(b) What are the most common thicknesses of the NaI(Tl) detector used?
(c) A gamma camera detector with a 20-cm eld of view is used to image the
lungs, which ll 75% of the image. The camera is set to accumulate 450,000
counts. Calculate the information density.
6. In pulse-height analysis, a 20% window means 10% on either side of the photo-
peak. True or false?
7. Describe how the digital camera works.
8. What is the advantage of a digital camera over an analog camera (Anger type)?
9. Why does a solid-state camera not require a PM tube?
128
Suggested Readings
9 Gamma Camera
Anger HO.Scintillation camera. Rev Sci Instr. 1958; 29:27
Cherry SR, Sorensen JA, Phelps ME. Physics in Nuclear Medicine. 4
th
ed. Philadelphia:
W.B.Saunders; 2012.
Early P.Private communication, 2005.
Erickson J.Imaging systems. In: Harbert J, da Rocha AFG, eds. Textbook of Nuclear Medicine,
Volume I: Basic Science. Philadelphia: Lea & Febiger; 1984.
Rollo FD, ed. Nuclear Physics, Instrumentation, and Agents. St Louis: CV Mosby; 1977.
RR
gs
222
Performance ofGamma Camera
10
10.1 Performance Parameters ofGamma Camera
The quality and detail of an image obtained by gamma cameras are affected by
several parameters associated with these imaging systems. These parameters include
spatial resolution, sensitivity, uniformity, and contrast, and they are described here
in detail. A brief description of the quality control tests for gamma cameras is also
included.
10.1.1 Spatial Resolution
The spatial resolution of a gamma camera is a measure of the ability of the device
to faithfully reproduce the image of an object, thus clearly depicting the variations
in the distribution of radioactivity in the object (Erickson 1984). The spatial resolu-
tion of a gamma camera is empirically dened as the minimum distance between
two points in an image that can be detected by the system. The overall spatial resolution (Ro) of a gamma camera comprises three components, namely, intrinsic resolution (Ri) of the detection system, collimator resolution (Rg), and scatter resolution
(Rs), and is given by
RR
The smaller numerical values of Ro indicate better resolution and vice versa.
oi
(10.1)
10.1.1.1 Intrinsic Resolution
Intrinsic resolution, Ri, is the component of spatial resolution contributed by the
detector and associated electronics, and is a measure of how well an imaging device
can localize an event on the image. Intrinsic resolution arises primarily from the
statistical uctuations in pulse formation that have been discussed in the section
entitled Gamma Ray Spectrometry in Chap. 8. The statistical variations in the
© The Author(s), under exclusive license to Springer Science+Business Media, LLC,
part of Springer Nature 2025
G. B. Saha, Physics and Radiobiology of Nuclear Medicine,
https://doi.org/10.1007/978-1-0716-4816-2_10
129
130
R
bc
e
10 Performance ofGamma Camera
production of light photons after γ-ray interaction in the detector and variations in
the number of electrons emitted from the photocathode and dynodes in the photomultiplier (PM) tubes have signicant effects on the intrinsic resolution. In gamma
cameras, the X, Y positioning of the pulses is improved by increasing the number of
PM tubes, thus improving the intrinsic resolution. Also, PM tubes with greater
quantum efciency and their improved optical coupling to the detector for greater
light collection provide better intrinsic resolution.
Intrinsic resolution improves with higher γ-ray energy and deteriorates with
lower energy because greater statistical uctuations occur in the production of light
photons by lower energy photons and vice versa. For example, the 140-keV photons
99m
of
Tc produce almost twice as many light photons in the detector as the 69- to
80-keV photons of
201
Tl and thus result in better intrinsic resolution. However, there
is little improvement in intrinsic resolution with photon energy above 250 keV
because of multiple scattering of photons within the detector that can result in photoelectric absorption (see below). Intrinsic resolution improves with narrow PHA
window settings, because scattered radiations are avoided.
Multiple Compton scattering of a γ-ray photon followed by absorption of all
scattered photons in the detector causes uncertainty in the X, Y location of the original γ-ray interaction and makes the intrinsic resolution, and hence spatial resolution,
worse. This effect is worse with thicker detectors and high-energy photons
(>250keV) because of the increased chances of multiple scattering. For this reason,
thinner detectors (0.63–1.84cm) are used in gamma cameras.
Most modern cameras have intrinsic resolution of the order of 4-mm full width
at half maximum (FWHM) for 140-keV photons of
99m
Tc.
10.1.1.2 Collimator Resolution
Collimator resolution, also termed the geometric resolution (Rg), constitutes the
major part of the overall spatial resolution and primarily arises from the collimator
design. In general, collimator resolution is worse than intrinsic resolution. As
already mentioned in Chap. 9, there are four major collimators: parallel-hole, pinhole, converging, and diverging. Of these, parallel-hole collimators are most commonly used in nuclear medicine.
The different parameters of a typical parallel-hole collimator are shown in
Fig.10.1. The spatial resolution for this collimator is given by the geometric radius
of acceptance, R
:
g
dt
e
g
t
(10.2)
where d is the hole diameter of the collimator, b is the distance between the collimator face and the source of radiation, c is the distance between the back face of the
collimator and the midplane of the detector, and te is the effective length of the collimator holes. The te is empirically given by te = t−2μ−1, where μ is the linear
attenuation coefcient of the photons in the collimator material (e.g., lead), and t is
COLLIM
DETECTOR
b
c
10.1 Performance Parameters ofGamma Camera
131
Rg
ATOR
ad
Source
Fig. 10.1 A parallel-hole collimator with thickness t, hole diameter d, septal thickness a, and
source-to-collimator distance b. The collimator is attached to a detector whose midplane is at a
distance
Table 10.1
features of parallel-hole
collimators on their
performance
Effect of various
Increasing Resolution Sensitivity
Number of holes ↔ ↑
Hole diameter ↓ ↑
Hole length ↑ ↓
Septal thickness ↑ ↓
Source-to-collimator distance ↓ ↔
t
the length or thickness of the collimator hole. This corrects for the penetration of the
two corners of the holes by the photons.
As seen from Eq. (10.2), the collimator resolution is improved by increasing the
length, t, of the collimator holes or by decreasing the diameter, d, of the holes. Thus,
long narrow holes provide better spatial resolution. Also, the collimator resolution
deteriorates with increasing source-to-collimator distance, b, and is best at the collimator face. Therefore, in nuclear medicine studies, patients should be placed as
close to the collimator as possible to provide the best resolution. The effects of various features of parallel-hole collimators on spatial resolution and sensitivity are
summarized in Table10.1.
The thickness a between the holes is called the septum. Septal penetration of
γ-rays plays an important role in the collimator resolution. High-energy photons
from outside the eld of view can cross the septum and yet interact in the detector,
thus obscuring the image. In the collimator design, a primary consideration is to
have negligible penetration by these extraneous photons through the septum to
reach the detector. However, it is practically impossible to stop all photons from
132
63dt/
10 Performance ofGamma Camera
penetration with any reasonable amount of material without substantial loss of
counting efciency. As a trade-off between penetration and collimator efciency, a
compromise value of 5% penetration is accepted, and the minimum septal thickness
a of a parallel-hole collimator can be calculated as
e
/
(10.3)
Septal penetration depends on the atomic number Z of the collimator material
and is low in high Z material. For cost-effectiveness, lead is commonly preferred for
use in the septa. It also depends greatly on the photon energy, and so gamma rays of
only ∼50–300 keV are suitable for present-day collimators, the most preferable
photon energy being 150keV.At energies below ∼50keV, photons are absorbed in
the body tissue, whereas at energies above ∼300keV, septal penetration of the photons can occur. Current collimators are made with appropriate septal thickness for
specic photon energies in order to limit septal penetration. Collimators are classied as low-energy collimators with a few tenths of a millimeter septal thickness (for
up to 140keV γ-rays) and medium-energy collimators with a few millimeters thickness (up to 360keV photons) (Cherry etal. 2012). Very high-energy collimators
also are available for counting 511keV photons. It is understandable that for a collimator of given diameter, the number of holes is greater in low-energy collimators
than in high-energy collimators. Various properties of different parallel-hole collimators are given in Table10.2.
In another classication, collimators are termed high-sensitivity and high- resolution collimators. Often, these collimators are made with an identical number of
holes with identical diameters, but with different thicknesses. Therefore, the highresolution collimators are made with longer holes and the high-sensitivity collimators with shorter holes. The spatial resolution for the high-sensitivity collimator
deteriorates sharply with the source-to-collimator distance. Low energy all purpose
Table 10.2 Various parallel-hole collimators and their features and properties
Hole
diameter
Collimator type
Low energy all
purpose
(LEAP)
Low energy
high resolution
(LEHR)
Medium energy 3.02 40.6 1.1–1.4 12.1 288 ∼280
High energy 4.32 62.8 1.3–3.0 13.8 176 ∼360
Ultra-high
energy
a
At 10cm from the collimator face
b
Adapted with permission from Halama J.Quality assurance in gamma camera and SPECT sys-
tems. www.medphysicwisc.edu/courses
(mm)
1.43 23.6 0.2 9.1 360 ∼140
1.11 23.6 0.3 7.5 230 ∼140
3.4 75.0 3.0–4.0 10.4 60 ∼511
Hole
length
(mm)
Septal
thickness
(mm)
Geometric
resolution
(mm)
b
a
Sensitivity
(cpm/μCi)
Optimum
energy
(keV)
20
20
COLLIMATOR-SOURCE DISTANCE
(cm)
A
(
10.1 Performance Parameters ofGamma Camera
133
(LEAP) are commonly used in nuclear medicine studies with collimators are
designed with intermediate values of resolution and sensitivity.
The geometric resolution for pinhole, diverging and converging collimators is
expressed by similar but somewhat complex equations, and their details are available from reference books on nuclear physics and instrumentation. In the case of
pinhole collimators, the image is magnied and the magnication depends on the
ratio of the hole-detector distance to the hole-object distance. The resolution varies
over the area of the object along with distortion of the image and the sensitivity falls
off with increasing distance from the collimator face. For converging collimators,
the image is magnied and the magnication increases with the distance from the
collimator face resulting in the deterioration of resolution and an increase in sensitivity, but the image is distorted. For diverging collimators, the situation is opposite
to that of the converging collimators. Having the collimator holes diverging away
from the detector results in minication of the object to t into a smaller detector.
So distortion occurs, and resolution and sensitivity vary over the object. The fan
beam collimator is basically a converging collimator and commonly used for cardiac and brain imaging. It gives better spatial resolution with poorer sensitivity than
parallel-hole collimators. The overall system resolutions with different collimators
are illustrated in Fig.10.2.
10.1.1.3 Scatter Resolution
Radiations are scattered by interaction with tissue in patients and with the detector.
It is possible that some of these radiations are scattered without much loss of energy
and fall within the eld of view, resulting in pulses of amplitude acceptable within
Fig. 10.2 Effect of
source-to-collimator
distance on overall system
resolution for various types
of collimators. A High
sensitivity parallel-hole. B
Diverging. C All-purpose
parallel-hole. D
Converging. E High
resolution parallel-hole. F
Pinhole. (From Rollo and
Harris 1977:407. Modied
from Moyer RA.J Nucl
Med 1974, 15:59)
16
(
12
8
4
0
0 5 10 15
B
C
D
E
F