Purpose: The importance of robust proton treatment planning to mitigate the impact of uncertainty is well understood. However, the computational cost grows with the number of uncertainty scenarios, which prolongs the treatment planning process. We developed a fast and scalable distributed optimization platform that parallelizes this computation over the scenarios.
Methods: We modeled the robust proton treatment planning problem as a weighted least-squares problem. To solve it, we employed an optimization technique called the Alternating Direction Method of Multipliers with Barzilai-Borwein step size (ADMM-BB). This is an iterative algorithm that splits the treatment planning problem into smaller subproblems, one for each proton therapy uncertainty scenario. The subproblems can be solved in parallel, allowing the distribution of computational load across multiple processors (e.g., CPU threads/cores). We evaluated ADMM-BB on 5 head-and-neck proton therapy cases, each with 13 scenarios accounting for 3 mm setup and 3.5% range uncertainties. We then compared the performance of ADMM-BB with projected gradient descent (PGD) applied to the same problem.
Results: For each case, ADMM-BB generated robust proton plans that satisfied all clinical criteria with comparable or better dosimetric quality than those generated by PGD. However, ADMM-BB's total runtime averaged about 5 to 8 times faster. This speedup increased with the size of the problem and the number of scenarios.
Conclusion: ADMM-BB is a powerful distributed optimization method that leverages parallel processing platforms, such as multi-core CPUs, GPUs, and cloud servers, to speed up the computationally intensive work of robust proton treatment planning. This results in 1) a shorter treatment planning process and 2) the ability to consider more uncertainty scenarios, which improves plan quality.
Protons, Treatment Planning, Optimization
TH- External Beam- Particle/high LET therapy: Proton therapy – dose optimization