A photon source model based on particle transport in a parameterized accelerator structure for Monte Carlo dose...

Area of Science:

• Medical Physics
• Radiation Oncology
• Computational Dosimetry

Background:

• Accurate modeling of medical linear accelerators (linacs) is crucial for precise Monte Carlo (MC) dose calculations in radiation therapy.
• Existing methods for determining linac photon spectra often lack realism, impacting the accuracy of dose simulations.

Purpose of the Study:

• To develop and validate an analytical photon source model for medical linacs.
• To achieve a more realistic determination of linac photon spectra using particle transport in parameterized accelerator structures.
• To improve the accuracy of MC-based dose calculations.

Main Methods:

• Designed primary and secondary photon sources based on photons attenuated/scattered by a parameterized flattening filter.
• Introduced parameters for primary photon fluence in penumbral regions.
• Optimized model parameters using calculated dose curves in water.
• Validated the model against Varian TrueBeam 6 and 15 MV phase space data (PSD) using EGS5 MC code.
• Performed 1D and 2D gamma analysis (1%/1 mm criteria) for dose distributions in water and lung phantoms.

Main Results:

• The optimized model accurately reproduced reference PSD spectral curves on- and off-axis.
• Depth dose and lateral dose profiles showed good agreement with reference PSD.
• Achieved 100% passing rates for 1D gamma analysis (1%/1 mm) across various field sizes and depths.
• Attained 100% passing rates for 2D gamma analysis in a heterogeneous lung phantom for 6 and 15 MV beams.
• Model optimization time was under 4 minutes.

Conclusions:

• The proposed analytical source model accurately generates photon fluence spectra from linacs.
• The model relies on physical properties and requires minimal optimization time, without needing detailed linac head geometry.
• This approach enhances the accuracy of MC dose calculations in radiation therapy.