Monte Carlo Particle Transport in Radiotherapy: Mice to Humans
Paul Keall and Joseph Perl
Stanford Medical School and SLAC, respectively
In radiation therapy, accurate estimation of dose to the tumor and surrounding structures is critical to delivering a useful treatment, with control of the tumor balanced against complications to normal tissues. Several techniques have been used to make these dose estimates, with increasing physics complexity being built into the models. But accurate calculations challenge current computational ability due to large variations of density and materials in human (and animal) anatomy.
The most accurate method, Monte Carlo particle transport, has many interesting roots at SLAC. The most popular Monte Carlo code for electron and X-ray therapy, EGSnrc, is based on the EGS code originally developed at SLAC in the 1970s, while a code used for newer proton and ion therapies, Geant4, has a development group active at SLAC today.
Current challenges in Monte Carlo modeling for radiotherapy include determining accurate models of the phase space and the treatment head geometry, reducing computation time, managing residual statistical uncertainty and improving ease-of-use. This presentation will give a background to Monte Carlo in radiotherapy and describe some of the ongoing Monte Carlo projects in small animal radiotherapy, human radiotherapy and design of next-generation treatment systems.
Paul Keall is associate professor in the Department of Radiation Oncology, Stanford School of Medicine. Joseph Perl is a research software developer at SLAC.