Energy Matrix Based Energy Layer Optimization for Proton Arc Therapy
G Zhang1, H Shen1,2*, Y Lin2, R Chen2, Y Long1, H Gao2*, (1) Shanghai Jiao Tong University, Shanghai, China (2) University of Kansas Medical Center, Kansas City, KS
Presentations
TU-F115-IePD-F5-1 (Tuesday, 7/12/2022) 1:15 PM - 1:45 PM [Eastern Time (GMT-4)]
Exhibit Hall | Forum 5
Purpose: Spot-scanning arc therapy (SPArc) is an emerging proton modality that can potentially offer a combination of advantages in plan quality and delivery efficiency, compared to traditional IMPT of a few beam angles. Unlike IMPT, frequent low-to-high energy layer switching (so-called switch-up (SU)) can degrade delivery efficiency for SPArc. However, it is a tradeoff between the minimization of SU times and the optimization of plan quality. This work will consider the energy layer optimization (ELO) problem for SPArc and develop a new ELO method via energy matrix (EM) regularization to improve plan quality and delivery efficiency.
Methods: The major innovation of EM method for ELO is to design an energy matrix that encourages desirable energy-layer map with minimal SU during SPArc, and then incorporate this energy matrix into the SPArc treatment planning to simultaneously minimize the number of SU and optimize plan quality. The EM method is solved by the fast iterative shrinkage-thresholding algorithm and validated in comparison with a state-of-the-art method, so-called energy sequencing (ES).
Results: EM is validated and compared with ES using representative clinical cases. For all cases, in terms of delivery efficiency, EM had fewer SU than ES; in terms of plan quality, compared to ES, EM had generally better target dose conformality as quantified by higher conformity index value, and better sparing of normal tissues as measured by lower mean dose to organs-at-risk and lower integral dose for body. Moreover, in terms of computational efficiency, EM was substantially more efficient than ES, and took only about 1/10 of computational time.
Conclusion: We have developed a new ELO method, namely EM, for SPArc using energy matrix regularization, which was shown to outperform ES (state-of-the-art), in both delivery efficiency and plan quality, with substantially reduced computational time.
Funding Support, Disclosures, and Conflict of Interest: This research is partially supported by the NIH Grant No. R37CA250921 and a KUCC physicist-scientist recruiting grant.
Keywords
Protons, Treatment Planning, Optimization
Taxonomy
Contact Email