Optimizing dose rate to minimize delivery time in proton pencil beam dose-driven continuous scanning

Area of Science:

• Medical Physics
• Radiation Oncology
• Computational Biology

Background:

• Dose-driven continuous scanning (DDCS) improves proton therapy efficiency over discrete spot scanning (DSS).
• Current DDCS optimization methods face limitations in computational efficiency and robustness due to fixed beam intensity and complex spot generation.

Purpose of the Study:

• Introduce a Break Spot-Guided (BSG) method with acceleration strategies (dose rate skipping, bounding) to optimize beam intensity and minimize beam delivery time (BDT).
• Propose a $\sigma$ -method for efficient move spot generation, balancing computational speed and dose accuracy.

Main Methods:

• Transformed the mixed-integer nonlinear programming (MINLP) problem into Traveling Salesman Problem (TSP) evaluations using break-spot dose rates.
• Integrated dose rate skipping and bounding strategies to reduce computational complexity.
• Employed the $\sigma$ -method for efficient move spot generation without compromising dose fidelity.

Main Results:

• The BSG-S-B method achieved a 64% reduction in BDT compared to DSS and a 57% reduction compared to a previous method in a prostate case.
• Acceleration strategies reduced computation time from hours to minutes (e.g., 10525.7s to 108.5s for prostate).
• The $\sigma$ -method reduced move spots by over 50% while maintaining dosimetric quality.

Conclusions:

• The BSG framework enhances computational efficiency, enabling clinical feasibility of scan path optimization with dose rate selection for DDCS.
• The $\sigma$ -method effectively decreases computational load and maintains dose accuracy, showing promise for proton therapy planning.