Evaluating X-Ray Source Arrays for Tomography and Tomosynthesis in Lung Cancer Radiotherapy
Presentations
TH-C-201-3 (Thursday, 7/14/2022) 10:00 AM - 11:00 AM [Eastern Time (GMT-4)]
Room 201
Purpose: Standard imaging in radiotherapy systems use a single kV x-ray source and flat panel detector. In this study we investigate adding novel carbon nanotube x-ray source arrays to otherwise standard systems to reduce pre-treatment scan times and enable real time volumetric imaging.
Methods: This study was performed using the Reconstruction Tool Kit (RTK) and the eXtended Cardiac And Thoracic (XCAT) synthetic phantom to simulate conventional and source array acquisitions of a patient before and during treatment. We investigated 4 configurations: single source single detector, source array single detector, conventional source with source array and 2 detectors, and 2 source arrays 2 detectors. Acquisitions were reconstructed using the Feldkamp-Davis-Kress (FDK) ,Total Variation with Alternating Direction Method of Multipliers (TVADMM) and the Prior Image Constrained Compressed Sensing (PICCS) algorithms. Image quality was quantified using Root-Mean-Square-Error (RMSE) over a central region of interest (ROI) and along tumour pixel profiles in each axis to quantify how precisely tumour position was captured.
Results: In pre-treatment imaging the source array geometry increased field of view size 75% and reduced scan time by 50-90% but with 27% worse RMSE relative to conventional source imaging at equivalent flux. Intra-treatment volumetric source array imaging with 1 second acquisition time had only 31% worse RMSE and 36% worse SSIM relative to the gold standard breath hold conventional source CBCT image.
Conclusion: This study demonstrates the trade-offs between conventional source and source array imaging on radiotherapy systems. Source array imaging widens the field of view, reduces scan time and enables volumetric imaging during treatment, but requires increased flux to achieve image quality equivalent to conventional single source imaging systems.
Funding Support, Disclosures, and Conflict of Interest: This research was supported by NHMRC project grant #1138899 and grant #1123068 which was awarded through the Priority-driven Collaborative Cancer Research Scheme and funded by Cancer Australia. Ricky O'Brien acknowledges support of a Cancer Institute NSW Career Development fellowship.
Keywords
Cone-beam CT, Tomosynthesis, Reconstruction
Taxonomy
IM- Cone Beam CT: Development (New Technology and Techniques)
Contact Email