Purpose: Computed tomography (CT) attenuation correction of myocardial perfusion in single-photon emission computed tomography (SPECT) /CT systems is possibility of misregistration between emission and transmission scans. This study aimed to evaluate the influence of misregistration using a polar map of 17 segments model.
Methods: Using the fusion software, we assessed the magnitude and direction of misregistration in 200 consecutive myocardial perfusion SPECT images with Technetium (Tc) tetrofosmin. After registration, CT data was shifted by ±1, ±2, and ±3 pixels along the cephalad/caudal, dorsal/ventral, and left/right axes, respectively. The registered image was compared with the shifted image.
Results: Misregistration between the SPECT and CT images occurred by 1-2 pixels in 127 cases (63.5%) and by 2 or more pixels in four cases (2%); the maximum misregistration was 1.2±0.4 pixels on average. The polar map scoring was most significantly affected by 3 pixel ventral shift. A ventral shift of 1 pixel affected the scores for the anterolateral and inferolateral segments, whereas a caudal shift of 1 pixel affected the scores for the anterior segment.
Conclusion: Since the 17 segments model can evaluate the position more precisely than the five segments model, it is possible to evaluate up to 1 pixel misregistration.
Download full-text PDF |
Source |
---|---|
http://dx.doi.org/10.6009/jjrt.2019_JSRT_75.9.892 | DOI Listing |
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!