Room: AAPM ePoster Library
Purpose: To develop and benchmark a 0.35T MR-guided linear accelerator (MR linac) head model in GEANT4 Monte Carlo code for a 6 MV flattening filter free (FFF) photon beam.
Methods: Geometric and energy parameters of the incident electron beam were tuned to match the measured beam profiles and percent depth dose (PDD) curves for the 6 MV FFF photon beam. The monoenergetic incident electron beam energy was iteratively optimized to match measured PDD data for a 3.3x3.3 cm² field size using a gamma index analysis. The magnitude of the Gaussian energy spread, spot size and angular divergence of the incident electron beam were changed iteratively to match measured PDD data and beam profiles at 5 cm depth for a 24.1x24.1 cm² field size using a gamma index analysis. A 2%/2mm gamma criterion was used for all comparisons of simulated and measured data.
Results: The optimal mean electron beam energy was found to be 6.0 MeV with a Gaussian energy spread of 1.50 MeV. The simulated PDD curves matched the measured curves with a passing rate of 100% for both 3.3x3.3 cm² and 24.1x24.1 cm² field sizes. The optimal electron beam spot size and angular divergence values were determined to be 1.0 mm and 0°, respectively. The simulated crossline beam profile matched the measured profile with a passing rate of 100% for the 24.1x24.1 cm² field size.
Conclusion: A vendor-independent GEANT4-based Monte Carlo model has been developed and benchmarked for a 0.35T MR linac. Good agreement was obtained between simulated and measured PDDs and dose profiles. In future studies, this model will be employed to investigate responses of several detectors in the presence of a 0.35T magnetic field.