Paramest Wongsa. Quantification study in myocardial phantom imaging by SPECT and SPECTCT : comparative study. Master's Degree(Radiological Science). Mahidol University. Mahidol University Library and Knowledge Center. : Mahidol University, 2009.
Quantification study in myocardial phantom imaging by SPECT and SPECTCT : comparative study
Abstract:
SPECT data acquisition is important for diagnostic information, according to
its ability to reflect the amount of radioactivity within the corresponding the region of
interest. However, there still may be errors in assigning activity due to the
characteristic resolution volume of the SPECT system. The loss in apparent activity
that occurs when an object partially occupies the sensitive volume of the imaging
instrument is called the partial volume effect. This work aimed to perform a
comparative quantification study in Myocardial Phantom Imaging by SPECT and
SPECT/CT. Three types of phantom (SPECT Phantom, NEMA IEC Body Phantom
setTM, and Anthropomorphic Torso PhantomTM with Cardiac InsertTM) were used to
calculate and compare defect sizes resulting from axial image reconstruction.The
iterative reconstruction with and without attenuation correction technique
(IRAC&IRNC) were applied. There was no considerable defect size differences in
NEMA IEC Body Phantom setTM, and SPECT Phantom study, within p< 0.05. There
was high correlation between the average threshold level and the defect size, which
revealed the more precise defect size corresponding to the higher threshold level. The
appropriate threshold level was found at 28-36 %, and a subsequent study was
performed by using Anthropomorphic Torso PhantomTM with Cardiac InsertTM;
threshold levels settings were 28, 32, and 36 %. The results of the differences between
maximum and minimum wall thickness (SD) calculation using the IRAC technique
were markedly increased in accuracy; 0.48 (0.23), 1.5 (0.62), and 0.48 (0.20) mm.
While a similar study using the IRNC technique resulted as follows: 3.50 (1.23), 2.93
(1.49), and 2.65 (1.16) mm., respectively. However, the optimal threshold level for
each defect size could not be determined, whereas the individual defect size changed