Scatter feature of a positron imager with dual plane geometry

Dedicated positron imagers with dual plane detectors for breast get high scatter fraction (SF) due to 3D data acquisition. In this study, we investigate the contribution of scatter from each breast and body outside of FOV, and its effects on lesion visibility. Beam stoppers (BSs) were designed to estimate the scatter distribution. All the data in this study were simulated using GATE. The FOV of this system is 196.8 mm×98.4 mm and the crystal element is 1.64×1.64×10 mm
³
LYSO. A XCAT phantom with F-18-FDG was simulated to obtain the body-scatter contribution to the left/right breast imaging individually. Besides, the breast phantom without body was simulated to obtain the scatter contribution from each breast. To mimic breast, three lesions with 3, 5, 8 mm diameters were placed inside a 100×100×80 mm
³
box phantom. Various T/B ratios were simulated for lesion visibility. Three BSs made of lead were placed between the upper detector and the box phantom. The phantom was scanned twice (w/wo BS). The BS blocks partially the recording of the primary events without affecting the scatter and random. From the counts of two scans, we can estimate the scatter distribution. The results of XCAT phantom show that the SF of left (right) breast imaging was 33.3% (33.1%) and 30.2% (29.8%), respectively, with and without the inclusion of body. The SF of box phantom was 36.4% which was higher than XCAT phantom due to the larger volume. At the T/B ratio 3:1 condition, the 3 mm lesion was faintly visible suffered from scatter. The scatter distribution estimated by BS was shown. The results indicate that left and right breast imaging are suffered almost the same scatter effect. The scatter from out of FOV could be ignored. The results of SF values and lesion visibility study show the scatter effect of positron imager with dual plane geometry could not be ignored. The BSs were designed to estimate the scatter distribution, and could be used in scatter correction in the future.