Optimization of Target Design and Dose Rates for FLASH-RT of An 18 MV Medical Linear Accelerator Model Via GEANT4 Monte Carlo Simulation
A Li1*, S.V. Kutsaev2, V Yu2, S Boucher2, K Sheng1, (1) UCLA Department of Radiation Oncology, Los Angeles, CA, David Geffen School of Medicine at UCLA (2) RadiaBeam, Santa Monica, CA
Presentations
WE-G-206-6 (Wednesday, 7/13/2022) 2:45 PM - 3:45 PM [Eastern Time (GMT-4)]
Room 206
Purpose: ROAD FLASH proposes a roadmap towards X-ray-based rotational intensity-modulated radiotherapy. To meet the FLASH dose rate requirement with X-rays, a high powered linac and an X-ray converter optimized for FLASH are needed. In this study, we optimize the target design specifications for an 18 MV source to reach the dose rates necessary for FLASH radiotherapy using GEANT4 Monte Carlo simulations.
Methods: The design of an 18 MV linear accelerator was optimized to maximize electron current. A phase space file was created for subsequent Monte Carlo simulation via GEANT4. The X-ray converter was a transmissive tungsten target. The electron leakage through the X-ray converter was reduced by a copper filter. The simulation modeled the linac beam path and was run with varying tungsten target and copper filter thicknesses. The X-ray and electron dose was calculated in a voxelized water phantom at 1 m source-to-surface distance. The maximized dose rate target and phase space configuration were further characterized by depth dose curves and lateral off-axis dose distributions corresponding to electron or photon contributions.
Results: The optimal thickness of an unfiltered tungsten target for maximizing dose rate was 0.6 mm, which resulted in a combined X-ray and electron dose rate of 7.94 Gy/s at 10 cm depth from 100 cm SSD. The electron contamination is reduced by a thousand fold with a 5 mm copper filter, resulting in a clean X-ray dose rate of 5.87 Gy/s at 10 cm depth.
Conclusion: This study determined the optimal target design specifications for producing the high dose rates. The dose rate can be further increased to meet the FLASH requirement with linac optimization and reduced SSD without affecting the optimized target configuration. Additionally, the individual contributions of the electrons and photons were characterized for possible mixed beam FLASH treatment planning.
Funding Support, Disclosures, and Conflict of Interest: This project is funded by NIH, award number NIH R01CA255432.
Keywords
Monte Carlo, Linear Accelerator, X-ray Production
Taxonomy
TH- External Beam- Photons: Computational dosimetry engines- Monte Carlo
Contact Email