(Tuesday, 7/14/2020) 3:30 PM - 5:30 PM [Eastern Time (GMT-4)]
Room: Track 3
Purpose:
We present a fully automated method for robust proton treatment planning using constrained, hierarchical optimization. To fill the gap between two common extreme robust optimization approaches, stochastic and worst-case, an approach based on the p-norm is used whereby a single parameter, p, can control the robustness-level in an intuitive way.
Methods:
We extend our in-house optimizer ECHO to include robust optimization. Thirteen scenarios accounting for setup- and range-uncertainties are included, and the maximum/mean/dose-volume constraints on organs-at-risk and target are fulfilled in all scenarios. We propose combining the objective-functions of the individual scenarios using the p-norm. The p-norm with p=1 results in the stochastic approach, and for large p-values (p~20) it resembles the worst-case approach, however, without focusing only on the worst scenario; intermediate robustness-levels are obtained by p-values in-between. The proposed approach is evaluated on three head-and-neck patients and an optimization phantom with different parameters, p={1,2,5,10,20}, and compared against the worst-case approach.
Results:
The robust plans meet all dose-constraints (maximum/mean/dose-volume constraints) for all scenarios. The median target-coverage (D98%) is reduced by 4 percent-points going from p=1 towards the worst-case approach. In contrast, the spread in the objective-function values is largest for the stochastic approach (p=1) and decreases with increasing p. Compared to the worst-case approach, the p-norm approach results in DVH bands for the clinical-target-volume which are closer to the prescription dose at a negligible cost for the worst scenario, improving the overall plan quality. On average, going from the worst-case approach to the p-norm approach with p=20, the median objective-function value over the scenarios is improved by 15% while the objective-function value for the worst scenario is only degraded by 3%.
Conclusion:
An automated robust treatment planning approach for proton therapy is developed, with the ability to control the robustness-level to fit the priorities deemed most important.
Funding Support, Disclosures, and Conflict of Interest: This work was partially supported by MSK Cancer Center Support Grant/Core Grant from the NIH (P30 CA008748).
Keywords
Protons, Optimization, Inverse Planning
Taxonomy
TH- External Beam- Particle/high LET therapy: Proton therapy – dose optimization
Contact Email
