WE‐C‐BRB‐09: Development of a GPU‐Based Monte Carlo Dose Calculation Package for Proton Radiotherapy

Purpose: Monte Carlo (MC) simulation is considered to be the most accurate method for proton therapy dose calculation. However, the long computation time limits it from routine clinical applications. Recently, graphics processing units (GPUs) have been widely used to accelerate computationally intensive tasks in radiotherapy. The purpose of this project is to develop a fast MC dose calculation package, gPMC, for proton dose calculation on a GPU. Methods: gPMC supports proton transport in the energy range of 0.5∼300 MeV. Proton transport is modeled by the Class II condensed history simulation scheme with continuous slowing down approximation. Ionization, elastic and inelastic proton nucleus interactions are considered. Energy straggling and multiple scattering are modeled. Secondary electrons and charged nuclear fragments are not transported and their energies are locally deposited. Secondary protons are stored in a stack and transported after the primary protons finish, while secondary neutral particles are neglected. gPMC is implemented on GPU under the CUDA platform with each GPU thread responsible for the transport of a single proton. A high performance random number generator and hardware linear interpolation are utilized. Results: We have validated gPMC in various cases including homogeneous phantoms with different materials and a phantom of a half slab geometry. A mono‐energetic mono‐directional proton beam is used. Good agreements between results from gPMC and Geant4 are observed, and the gamma passing rate for 3%/3mm criterion is over 99.9% in the region with dose greater than 10% maximum dose. It takes 6.3∼24.1 sec to simulate 10 million source protons depending on the phantoms and the energy. A high efficiency has been achieved compared to the computational time of tens of CPU hours for Geant4. Conclusions: We have developed a GPU‐based MC proton dose calculation package, gPMC. The achieved accuracy and efficiency is promising for clinical applications.