Gate for brachytherapy applications
Résumé
GATE is a simulation toolkit for PET and SPECT applications based on the GEANT4 code. Specific modules are added on top of GEANT4, for PET and SPECT requirements and facilitate the usage of the code. Our study shows that Gate can also be applied to model brachytherapy-radiotherapy applications. In this approach GATE platform is used to calculate the relevant dosimetric quantities for treatment planning in brachytherapy-radiotherapy. We are trying to validate the GATE code (low energy range for photons and electrons) by simulating 106Ru/106Rh source and all dosimetric quantities recommended by the AAPM Task Group 43 (TG-43) for three designs of 125I seeds (Symmetra Model I25.S06, Best Model 2301 and model 6711).We study also the analysis methods for dose calculations adapted for the dosimetric applications. The analysis method will be implemented in GATE simulation platform. Thus to calculate air kerma strength, we used the track length estimator method. We modified GATE toolkit to recover information concerning the track energy. All the sources characteristics have been simulated respecting welded ends, radioactive distribution in the source, materials and rays energy spectra. Auto-absorption of the sources as titanium characteristic X-ray production has been taken into account. Our Monte Carlo results were compared against consensus values reported in the AAPM TG-43 dosimetry protocol stemming from Monte Carlo simulations and measurements and other Monte Carlo codes. Concerning 106Ru/106Rh, we simulated CCB, CCA, CCX and CCZ applicators models manufactured by BEBIG. The results are compared with those calculated by the Hokkanen method empirical and obtained with MATHEMATICA software; they are also compared with those resulting from the Plaque Simulator (analytical toolkit) and with the experimental measurements obtained with scintillator plastic exit of the BEBIG certificate calibration. Results are satisfing. The total Monte Carlo uncertainty affecting our results is composed of evaluated statistical and cross sections uncertainties. We calculated in liquid water the dose rate function and anisotropy function for the three sources (125I) previously quoted. The calculated dose-rate constants for the three sources were found to be quite different from the recommended values of the TG-43 report, especially for the model 6711 and the Symmetra model. Radial dose functions and anisotropy functions are in very good agreement with recommended TG-43 values in particular for the Best Model 2301. An ocular brachytherapy treatment using thirteen 125I model 6711 seeds has been simulated and compared to analytic treatment planning. The use of GATE (Geant4) to model and calculate dosimetric quantities particularly for low-energy brachytherapy sources has been shown.