A Monte Carlo dose calculation algorithm for proton therapy

Area of Science:

• Medical Physics
• Computational Biology
• Radiotherapy

Background:

• Proton beam therapy requires accurate dose calculations for effective cancer treatment.
• Existing Monte Carlo (MC) simulation codes can be computationally intensive, limiting their clinical application.
• Optimizing MC algorithms is crucial for improving the speed and efficiency of treatment planning.

Purpose of the Study:

• To introduce VMCpro, a novel MC code designed for proton beam therapy treatment planning.
• To describe the implementation of electromagnetic and nuclear interactions within VMCpro.
• To evaluate the speed and accuracy of VMCpro compared to established MC codes.

Main Methods:

• VMCpro utilizes a Voxel Monte Carlo algorithm adapted for proton interactions in human tissue.
• It incorporates a Class II condensed history algorithm to model continuous energy loss, ionization, and scattering.
• Electromagnetic and nuclear interactions, including proton nucleus scattering and reactions, are implemented.

Main Results:

• VMCpro demonstrates significant speed improvements: 13x faster than FLUKA and 35x faster than GEANT4 in phantom simulations.
• Speed advantage increases in patient dose calculations due to weak dependency on grid heterogeneity.
• Dose distributions calculated by VMCpro show agreement with GEANT4, with deviations within 1% or 0.5 mm in high-gradient regions.

Conclusions:

• VMCpro offers a computationally efficient and accurate solution for proton beam therapy treatment planning.
• The code's speed and accuracy make it a viable tool for clinical applications.
• VMCpro has the potential to enhance the workflow and precision of proton therapy.