ABSTRACT
Dose calculation is at the center of radiation therapy planning. Its result serves as input to the optimization algorithm and is at the core of plan creation and evaluation. Dose calculation methods are diverse, going not only from fast to slow but also from inaccurate to accurate. Monte Carlo (MC) methods, now alongside the relatively recently introduced grid-based Boltzmann solvers, are at the “slow but accurate” extreme of the diversity spectrum of dose calculation methods used in radiation therapy [1]. e appearance of absolute parallelism, where the contributions of millions of individual particles are added to generate the nal result, also made it a seemingly ideal candidate for parallelization schemes.
