14 Determination of small MLC-fields output correction factors for four commercial detectors used in clinical dosimetry using GATE/Geant4 Monte Carlo simulations

Abstract

Introduction The aim of this work is to calculate detector-specific output correction factors using Monte Carlo (MC) simulation for small nominal multi-leaf collimator (MLC) photon beams, according to the recent TRS-483 protocol [1] , [2] . is determined for four commercially available PTW active detectors (T31016 and T31022 PinPoint 3D ion chambers, T60019 microDiamond and T60017 Ediode) used for small photon fields measurements. This study considers the 6WFF, 6FFF and 10FFF nominal beams of a Varian Truebeam STx linear accelerator. Methods The precise geometries and material compositions of the detectors are modeled in detail in the Geant4 applications for emission tomography (GATE) code using blueprints and data provided by PTW-Freiburg. The accelerator head was formerly modeled in GATE for nominal small MLC photon fields ranging from 0.5 x 0.5 cm2 to 3 x 3 cm2, providing phase-space files as input to this study. Absorbed dose scored over the active volumes of the detectors in water are compared to absorbed dose scored in a small reference water volume in the simulations. obtained by MC simulations are compared to those obtained by EBT3 Gafchromic films and the T60003 PTW diamond detector measurements, considered as benchmarks. Combination of simulated and experimental output ratios are also used to report for. Obtained MC values are then compared to data provided in the TRS-483. Results For the 6WFF, the absorbed dose scored in the detectors (T31016, T31022 and T60017) are in good agreement ( 2%) with the absorbed dose calculated in a small volume of water for the 1 x 1 cm2and 2 x 2 cm2 MLC openings. The resulting uncertainties for the simulations are estimated to be within 0.5% at all field sizes. A good agreement was also found with the EBT3 and diamond measurements, with a relative difference of 1% for field sizes greater than 0.5 x 0.5 cm2. In the worst case, for the 6WFF beam and 0.5 x 0.5 cm2 field size, the Ediode response differs from that of EBT3 by around 4%. The output ratio for field sizes smaller than 2 x 2 cm2 are overestimated by the T60017 diode and underestimated by the T31016 and T31022 ion chambers. This work shows an overall good agreement, within 1.5% relative difference with stated values in the TRS-483. Conclusions Simulations in GATE for the determination of demonstrate comparable results with values obtained with EBT3 film and T60003 diamond detector measurements. Results for the 6FFF and 10FFF and for the T60019 are ongoing.