Room: Exhibit Hall | Forum 4
Purpose: To significantly enhance multi-helix rotating shield brachytherapy (H-RSBT) treatment planning quality by solving convex optimization algorithm and selecting the delivery keyways (helices). The proposed method aims to maximize the dose delivery as well as to reduce the treatment time by using fewer keyways while guaranteeing planning performance.
Methods: A H-RSBT applicator design has the source/shield makes 54Â° of rotation per 5mm of its translation along the applicator, with 1.7mm spacing between dwell positions. Each of delivery path with a different starting shield opening orientation is a keyway. We devise the first-order convex optimization algorithm with consideration of minimizing treatment time of a plan, and solve the problem by the proximal graph solver (POGS). The potential 24 keyways are separated into smaller sectors, and we select keyways with high delivery time. The proposed keyway selection optimization method was validated using five clinical datasets of MRI-guided, cervical cancer HDR brachytherapy. Our treatment plans were compared with conventional H-RSBT with 6-equal-spacing keyways. Different number of keyways were tested in terms of treatment delivery time and dosimetric plan quality parameters. Considered regions-of-interests consisted of rectum, bladder, sigmoid colon, HR-CTV, and HR-CTV boundary in the asymmetric dose-volume optimization with smoothness control. Dose calculation resolution was 1 mm x 1 mm x 3 mm for all cases.
Results: 6-keyway selection approach outperformed current H-RSBT in both plan quality and treatment delivery time, on average 1 to 2 Gy higher in HR-CTV Dâ‚‰â‚€ along with superior values in HR-CTV Dâ‚?â‚€â‚€ and all organs-at-risk D2cc. Our method for selection of 3 to 5 keyways saved from 26 to 57% of treatment delivery time, while maintaining reasonable plan quality.
Conclusion: Keyway selection optimization provides superior treatment quality as well as considerably reduce its delivery time.
Funding Support, Disclosures, and Conflict of Interest: This research is partially supported by the National Institutes of Health (NIH) under grant number NIBIB R01EB020665