Fast optimization and dose calculation in scanned ion beam therapy.

Area of Science:

• Medical Physics
• Radiation Oncology
• Computational Biology

Background:

• Particle therapy (PT) offers advantages over photon irradiation for static tumors, including enhanced biological effectiveness and conformal dose shaping.
• However, PT's sensitivity to geometric changes complicates its use for moving organs, necessitating adaptive radiotherapy (ART) strategies.
• Treatment replanning (TRP), an ART strategy, involves re-optimizing treatment plans based on daily imaging before each fraction.

Purpose of the Study:

• To accelerate the time-consuming process of treatment plan optimization and dose calculation for particle therapy, particularly for scanned beam treatments.
• To implement and evaluate fast optimization and dose calculation routines within the GSI in-house treatment planning system (TRiP98) for enabling TRP.
• To improve the feasibility of adaptive particle therapy for moving targets through enhanced computational efficiency.

Main Methods:

• Conducted a code analysis to optimize calculation processes within the TRiP98 treatment planning system (TPS).
• Implemented routines to support parallel execution of the TPS code.
• Benchmarked the performance improvements by measuring calculation times for therapy treatment planning.

Main Results:

• Code optimization improved calculation times for treatment planning (optimization and dose calculation) by a factor of 10.
• Parallelization of the TPS achieved speedup factors of 12 (original version) and 5.5 (code-optimized version).
• The combined optimizations resulted in total speedup factors of up to 55 for the treatment planning system.

Conclusions:

• The enhanced treatment planning system (TPS) can now complete treatment planning for ion beam therapy, including prostate irradiation with organs at risk, in under 6 minutes.
• This significant speedup facilitates the practical implementation of adaptive radiotherapy (ART) strategies for particle therapy.
• The optimized TPS supports efficient treatment replanning (TRP), crucial for managing interfractional motion in particle therapy.