Q5. Briefly describe the principles of SPECT and PET imaging and discuss their respective advantages and disadvantages. 1. Single photon emission computed tomography (SPECT)

sunglowcitrineΠολεοδομικά Έργα

15 Νοε 2013 (πριν από 4 χρόνια και 6 μήνες)

109 εμφανίσεις

Q5. Briefly describe the principles of SPECT and PET
imaging and discuss their respective advantages and


Single photon emission computed tomography (SPECT)


Basic principles

Tomographic imaging attempts to depict the activity distribution
in a single
cross section of the patient. Computed tomography requires the acquisition of
a set of projection images from at least a 180° arc around the patient. These
images are then mathematically reconstructed to form cross section images.

Emission com
puted tomography performed with radionuclides emitting x

rays without angular correlation is called single photon emission
computed tomography (SPECT).

Almost all commercially available SPECT systems are based on the gamma
camera. A single gamma

camera head, mounted on a rotating gantry, is
sufficient to acquire the data needed for tomographic images. The gamma
camera acquires two
dimensional (2
D) projection images at equally spaced
angular intervals around the patient. These images provide the
D projection
data needed for reconstructing cross
sectional images for all axial locations
(slices) covered by the field
view (FOV) of the gamma camera.

The required number of views (angular projections) is generally between 64
and 128 to avoid strea
k artifacts.

Although data acquired over an arc of 180° are sufficient for tomographic
reconstruction when ideal projection images are available, a 360° rotation is
often used in SPECT to minimize errors due to attenuation and nonuniform


camera heads on older SPECT systems followed circular orbits around
the patient while acquiring images. To improve image spatial resolution by
keeping the camera head in close proximity to the surface of the body
throughout the acquisition, newer SPECT sy
stems provide an elliptical orbit or
even a "body contouring" orbit.

An important problem that deteriorates image quality in emission tomography
is the statistical noise. To increase the system sensitivity, thus reducing the
noise, modern SPECT systems ar
e generally using 2 or 3 camera heads,
allowing several angular projections to be acquired simultaneously.

The SPECT systems that are using rotating camera heads require an
acquisition time of several minutes in order to obtain a complete set of
ns. This long acquisition time prevents imaging of fast biologic
processes. Moreover, traditional collimator design limits reconstructed spatial
resolution to 1 cm or greater. Many efforts have been made to address this
limitations and some specialized sy
stems have been developed; however,
their clinical application is still limited.

A general limitation of emission imaging is the lack of anatomical information
on the images. Some success has been achieved using computer algorithms
to co
register emissi
on images with MRI or CT images that have been
acquired on separate imaging devices and provide excellent anatomic details.

However, a more direct approach is to acquire the images simultaneously on
a multimodality imaging system. Systems, called SPECT/CT

comprising a clinical SPECT camera attached to a clinical x
ray CT scanner
are now commercially available.

uniform resolution, patient attenuation and scattered radiation detection
distort the desired linear relationship between signal level

and amount of
activity. Therefore, somewhat modified data acquisition approaches and/or
data corrections must be applied.


Positron emission imaging (PET)


Basic principles

Positron emission tomography
(PET) requires the use of positron
clides. Images depicting the distributions of these radionuclides in
patients are generated.

The fundamental characteristic of positron emission tomography is the
coincidence detection of the two 511 keV annihilation photons that are emitted
when a positr
on annihilates with an electron from the tissues. The
simultaneous (or nearly simultaneous) detection of the 2 annihilation photons
in 2 detectors that are 180° apart, allows to localize their origin along the line
that connects the detectors,
without the
use of absorptive collimators

Coincidence logic circuitry is employed to analyze the signals from the
opposing detectors and a coincidence event is assumed to have occurred
when a pair of events is recorded within a specified
coincidence timing
ypically 6
12 ns). Although annihilation photons are emitted
simultaneously, a small but finite coincidence window width is needed to allow
for differences in signal transit times (cables, electronics) and for the different
distances covered by the 2 photo
ns from annihilation to detection.

Consequently, other types of events can be detected in the coincidence
window and must be taken into account (see disadvantages paragraph).

. Principle of electronic collimation.


using multiple opposing detectors in a complete ring (or other geometric
array) around the patient, coincidences between each detector and the
multiple detectors on the other side of the array can be recorded.

A stationary PET system allows to acquire da
ta for all projection angles
simultaneously. This greatly reduces acquisition duration compared to rotating

. Examples of recorded lines of response in a stationary PET detector ring.

In theory, it is possible

to determine the location at which annihilation took
place along the line of response by measuring the difference between the
arrival times of the 2 photons in the detectors. This technique, called
PET, would allow the formation of to
mographic images
mathematical reconstruction algorithms
. Unfortunately, the rise times of
light output from available scintillators are too slow to provide such a high
level of timing resolution. However, the development of new scintillator
ls has recently led to the first commercial TOF
PET system with a
timing resolution of about 600 ps, that is adequate to achieve localization to
within approximately 10 cm. Such a positioning accuracy does not improve
image spatial resolution but rather in
creases the signal
noise ratio by
constraining individual events within a smaller volume during the image
reconstruction process.

A number of corrections are also required in PET in order to get quantitative
images where intensity is proportional to th
e activity concentration at the
corresponding location in the object. This is necessary for accurate
comparisons of activity levels in different organs or between normal and
diseased tissues.

Radionuclides that are used in PET are isotopes of elements tha
t are
important constituents of all biologic substances (C, O, N, F), and they can be
used to label a wide variety of biologically relevant tracers.


PET advantages

Although the annihilation photons could be detected using SPECT
detectors, these systems
are not optimally designed for the relatively
high energy (511 keV) of annihilation photons. They have a rather low
detection efficiency at these energies and require special high
collimators that further decrease the overall efficiency. Moreover, t
directional characteristics of annihilation photons is better exploited
by special coincidence detection systems developed for PET.

annihilation coincidence detection
(ACD) actually defines the
volume from which the annihilation photons were emitte
d. Such an
"electronic collimation" is much less wasteful of photons than the
collimation by absorption used in SPECT.

Moreover, ACD avoids the spatial resolution degradation with
distance that is encountered when collimation is used to form
projection im

A stationary PET system that completely surrounds the patient allows
to acquire data for all projection angles simultaneously. This greatly
reduces acquisition duration compared to rotating systems (like those
used, for instance, in SPECT).

As i
t is the case with SPECT imaging, PET systems have been
integrated with x
ray CT technology to create combined PET/CT
scanners in a single gantry. These systems can produce fused
images showing both anatomy (CT) and function (PET) with near
perfect registr
ation of the images.

The overall sensitivity of a PET system for a small source located
near the center of the scanner ranges from 0.2 to 0.5% (0.002 à
0.005 cps/Bq) in 2
D acquisition mode and reaches between 2 and
10% (0.02 à 0.10 cps/Bq) in 3
D acquisi
tion mode. For comparison,
the sensitivity of a SPECT system with a general
purpose parallel
hole collimator is in the range of 0.01% to 0.03% depending on the
number of detector heads.


PET disadvantages

Because of their very short lifetimes, most of th
e positron
radionuclides used in PET must be prepared on site with a dedicated
biomedical cyclotron, leading to higher cost. An exception is fluorine
18 that is mainly used in the labeling of a glucose analog,
fluorodeoxyglucose (FDG), whose l
onger half
life allows production in
regional distribution centers and shipment to hospitals a few tens
kilometers away. FDG is the most widely used positron
radiopharmaceutical with a wide range of clinical applications in
cancer, in the heart an
d in brain.

Both random and scatter coincidences produce incorrect positional
information that leads to a relatively uniform background on the
image, resulting in a loss of contrast and distorting the relationship
between image intensity and actual activi


Many unscattered photons interact with the detectors by Compton
scattering and deposit less than 511 keV. Therefore, these valid
events are rejected by an energy window that only encompasses the
photopeak. Thus, the width of the energy window results

from a
compromise between scatter rejection (narrow window) and
sensitivity (wide window).

Even though the attenuation coefficient for 511 keV annihilation
photons in soft tissues is lower than for photons emitted by most
radionuclides used in SPECT, the

average path length for both
annihilation photons to escape is much longer. Thus attenuation is
more severe in PET than in SPECT. Attenuation causes a loss of
information and, because the loss is not the same for all lines of
response, it produces artifac
ts in the reconstructed images. In
SPECT, attenuation correction depends on the depth from which the
photon has been emitted (A); this is not the case in PET(B), where
attenuation correction can be obtained with an external source by
combining a blank scan

(C) and a transmission scan (D).