Exhibit Hall | Forum 5
Purpose: FLASH-RT can provide additional biological dose sparing for OAR, via ultra-high dose rate delivery, for which proton cyclotron system is a popular choice. Further increase of dose rate can be promised by higher beam current, which in turn is proportional to the minimum monitor unit (MMU) for cyclotron. However, the optimization with large MMU threshold is highly nonconvex, under which most current optimization algorithms cannot solve. This work aims to develop an effective FLASH optimization algorithm, so-called modified orthogonal matching pursuit (MOMP) that can handle large MMU threshold to further enhance dose rates.
Methods: During MOMP, we simultaneously optimize physical dose distribution and the FLASH coverage (i.e., the volume satisfying FLASH dose rate threshold and FLASH dose threshold). The determination of active sets for DVH planning constraints and the MMU constraints are decoupled using iterative convex relaxation method. To deal with MMU constraints, the non-zero spots are greedily selected by using orthogonal matching pursuit, and then we solve quadratic programming with respect to nonzero spots. During the iterations, the newly found non-zero spots are adaptively added or excluded based on planning objective values. On the other hand, since it is a tradeoff between physical dose coverage and biological FLASH coverage. after plan optimization FLASH effective dose is computed as a FLASH dose modifying factor (DMF) weighted physical dose distribution.
Results: Alternating direction method of multipliers (ADMM) is selected as a benchmark optimization method. MOMP was shown to substantially outperform ADMM, for FLASH optimization problem with large MMU thresholds, in terms of both target dose conformality (quantified by Dmax given D95=100%) and OAR sparing (quantified by FLASH effective dose).
Conclusion: MOMP is developed that can handle large MMU thresholds in order to further enhance dose rates and thus the FLASH coverage.
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.
Treatment Planning, Protons, Inverse Planning
TH- External Beam- Particle/high LET therapy: Proton therapy – dose optimization