(Sunday, 7/12/2020) [Eastern Time (GMT-4)]
Room: AAPM ePoster Library
Purpose: of CBCT deteriorates substantially for obese patients due to higher fluence of scattered radiation. While recently introduced 2D antiscatter grids (2D-ASG) can reduce scatter, a small fraction of scatter fluence is transmitted through the 2D-ASG, degrading the image quality. Hence, we propose a new method for improved scatter mitigation, where 2D-ASG was used as a scatter correction device, as well as a scatter rejection device. In this work, we compared the imaging performance of our method in the context of obese patient imaging.
Methods: septal shadows introduce a fluence modulation pattern in projections, and modulation pattern’s amplitude correlates with the scatter fluence. Our method exploited this correlation to measure scatter transmitted through the 2D-ASG. To validate this approach, scatter intensity measured with our method was compared to the conventional beam-stop technique. Furthermore, CBCT experiments were performed using both standard and obese phantoms with a linac-mounted CBCT and 2D-ASG. Image quality was evaluated with respect to clinical CBCT (Varian TrueBeam) and helical CT.
Results: intensity measured with our method and beam-stop method agreed. Our method exhibited better CT number accuracy, uniformity, and less artifacts than clinical CBCT. When phantom size was increased from standard to obese, HU values for bone and soft tissues degraded by 18% and 4% in clinical CBCT. With our method, average HU degradation for bone and soft tissues was only 6% and 2%. HU degradation in helical CT was 5% and 1% for bone and soft tissues.
Conclusion: unified scatter mitigation method provided CT number accuracy close to helical CT under very high scatter conditions observed during obese patient CBCT imaging. With implementation of other data correction methods, such as beam hardening, we predict that comparable CT number accuracy in helical CT and CBCT may be achieved with our proposed method for CBCT-guided radiation therapy.
Funding Support, Disclosures, and Conflict of Interest: This work is supported in part by grants from NIH/NCI (R21CA198462) and Cancer League of Colorado.
Keywords
Not Applicable / None Entered.
Taxonomy
Not Applicable / None Entered.
Contact Email
