Validation of the gate Monte Carlo code for radiation therapy and application to inhomogeneities

Introduction The treatment planning system (TPS), used in most cases, is based on fast and accurate analytical algorithms for dose estimation. However, limitations appear in inhomogeneous environments with strong gradient of density, as in the lungs. Monte Carlo solutions exist for complicated treatment with high inhomogeneity, such as the gold-standard BEAMnrc/EGSnrc. Nevertheless, the computation times are too long for routine use. In order to implement other Monte Carlo codes for external radiation therapy, our study focuses on the validation of GATE/ GEANT4. Materials and methods The linear medical accelerator CLINAC 2100C (Varian) from Belfort-Montbeliard hospital is modelised in GATE 6.2 for a 6 MV photon beam. The reference measurements are determined by two types of ionization chambers (PTW 31002 and PTW 31014) and are compared with the results of the both Monte Carlo codes (GATE and BEAMnrc). Once GATE has been validated, the performance of both codes will be evaluated in the inhomogeneities (dentures). Results The simulation parameters of GATE were adjusted to validate the accelerator modeling from percentage depth dose and profiles in the reference conditions. The results of the both codes with high density inhomogeneities are close. Conclusion The accelerator model was validated using GATE. The first analyses about the inhomogeneities give promising results. Thereafter, the evaluation of the code will go on more complex models (anthropomorphic phantom, breathing 4D phantom) always using BEAMnrc as a reference.