Room: AAPM ePoster Library
Purpose: At the dose rates common in proton therapy, Compton cameras (CC) suffer from high dead-time and false coincidence rates. The purpose of this study is to determine how the detector configuration can be optimized to alleviate these problems.
Methods: Using a Monte Carlo plus Detector Effects (MCDE) model of our CZT based CC, we modified several detector performance parameters: 1) event coincidence acceptance window temporal length; 2) dead time scaling; 3) individual vs grouped crystal readout; 4) reduced crystal thickness. For each configuration we assessed the false coincidence percentage, event classification for intra-module events, detection efficiency for inter-module events, and total amount of data suitable for image reconstruction. Eight detector configurations were tested at three dose rates for a 150 MeV proton pencil beam incident on HDPE: 0 kMU/min (reference), 20 kMU/min (minimum clinical dose rate) and 180 kMU/min (maximum clinical dose rate).
Results: Reducing the event coincidence acceptance window reduces the percentage of false coincidence data without changing the total amount of recorded data. Conversely, reduction in dead time increases the total counts recorded, but does not change false coincidence rates. Individual crystal readout outperformed grouped crystal readout in all aspects. Thinner detector crystals tend to have better data quality compared to the reference detector, although their total data output is significantly reduced due to lower interaction cross-section. Individual crystal readout with reduced dead time offered the highest amount of data suitable for reconstruction, while the thinnest crystal setup offered the least.
Conclusion: This study can be used to help guide future designs of CCs depending on the required data quantity, quality, or readout complexity of the system.
Funding Support, Disclosures, and Conflict of Interest: Funded by National Institutes of Health National Cancer Institute Award R01CA187416.