GPUMCD: a new GPU-oriented Monte Carlo dose calculation platform

摘要

Purpose: Monte Carlo methods are considered the gold standard for dosimetriccomputations in radiotherapy. Their execution time is however still an obstacleto the routine use of Monte Carlo packages in a clinical setting. To addressthis problem, a completely new, and designed from the ground up for the GPU,Monte Carlo dose calculation package for voxelized geometries is proposed:GPUMCD. Method : GPUMCD implements a coupled photon-electron Monte Carlosimulation for energies in the range 0.01 MeV to 20 MeV. An analogue simulationof photon interactions is used and a Class II condensed history method has beenimplemented for the simulation of electrons. A new GPU random number generator,some divergence reduction methods as well as other optimization strategies arealso described. GPUMCD was run on a NVIDIA GTX480 while single threadedimplementations of EGSnrc and DPM were run on an Intel Core i7 860. Results :Dosimetric results obtained with GPUMCD were compared to EGSnrc. In all but onetest case, 98% or more of all significant voxels passed a gamma criteria of2%-2mm. In terms of execution speed and efficiency, GPUMCD is more than 900times faster than EGSnrc and more than 200 times faster than DPM, a Monte Carlopackage aiming fast executions. Absolute execution times of less than 0.3 s arefound for the simulation of 1M electrons and 4M photons in water formonoenergetic beams of 15 MeV, including GPU-CPU memory transfers. Conclusion :GPUMCD, a new GPU-oriented Monte Carlo dose calculation platform, has beencompared to EGSnrc and DPM in terms of dosimetric results and execution speed.Its accuracy and speed make it an interesting solution for full Monte Carlodose calculation in radiation oncology.