Biomedical Engineering Reference
In-Depth Information
FIGURE 5.16: (See color insert.) Cardiac FDG PET/CT scan (long axis
view). Non-corrected PET images showing position of the heart during CT
(top: end-expiration; bottom: end-inspiration) and PET are shown on the
left side, attenuation-corrected PET images on the right. Note the apparent
uptake defect in the lateral wall (arrow) due to misregistration between end-
expiration CT and PET.
Several approaches to this potential drawback have been discussed in the
recent literature. One method proposed in cardiac PET/CT and SPECT/CT
is the manual alignment of emission and CT image data [63] [34] which however
requires user interaction; additionally, as respiratory motion is usually not
confined to pure translational motion, it is not clear to what extent the CT
data can be properly aligned to the PET data [36]. Another method would
be to try to improve the CT protocol regarding adapted breathing commands
for the patient to minimize transmission and emission misregistration [32].
Alternatively, one could aim to simulate the situation in stand-alone sys-
tems, i.e., acquiring CT data over one or several respiratory cycles. This can
be done either by averaging CT data sets taken at several different respiratory
steps in time (Cine CT) [3] [75] or by acquiring a very slow and long CT scan
at low tube current, thereby introducing the motion blur into the measured
transmission data [71] [82]. Both methods were shown to successfully reduce
misalignment between emission and transmission data.
Besides reducing misalignment effects, simulating a stand-alone system
obviously introduces motion effects into the transmission data which in the
case of large respiratory displacements may degrade image quality due to
image blurring. Thus, several methods aim to reduce the impact of respiratory
motion in emission tomography data. These approaches can be referred to as
motion compensation or motion correction methods. The basic idea here is to
use motion information acquired during the scan to divide the raw data into
several subsets (gates) of much reduced motion, comprising just one phase
of the respiratory cycle. This data can then be used to get near motion-free
 
Search WWH ::




Custom Search