Evaluation of a commercial MRI Linac based Monte Carlo dose calculation algorithm with GEANT4

Area of Science:

• Medical Physics
• Radiation Oncology
• Computational Dosimetry

Background:

• Accurate dose calculation is crucial for radiation therapy planning.
• Monte Carlo (MC) methods are advanced tools for simulating radiation transport and dose deposition.
• The integration of magnetic resonance imaging (MRI) with linear accelerators (LINACs) presents new challenges for dose calculation algorithms.

Purpose of the Study:

• To compare the accuracy of the GPUMCD (GPU Monte Carlo Dose) algorithm against GEANT4 for dose calculation.
• To evaluate the dosimetric impact of a 1.5 T external magnetic field on dose calculations.
• To assess the performance of GPUMCD for use in MRI-Linac treatment planning.

Main Methods:

• A standalone GPUMCD algorithm was compared with GEANT4 (v10.1).
• Simulations were performed in water, ICRU lung, and compact-bone phantoms, with and without a 1.5 T magnetic field.
• Dose calculations used 2 MeV photons and a 7 MV spectrum, with field sizes from 1.5x1.5 to 10x10 cm², scored in 1 mm³ voxels.

Main Results:

• GPUMCD showed excellent agreement with GEANT4, with dose differences <1% in water and <±2% in heterogeneous phantoms.
• Over 99% of voxels passed volumetric gamma analysis criteria (2%-2 mm).
• The 1.5 T magnetic field caused significant dose alterations at material interfaces, varying with material density and field size (e.g., up to 47.83% dose increase at water-lung interface for 5x5 cm²).

Conclusions:

• The GPUMCD algorithm demonstrates high accuracy comparable to GEANT4, even under a 1.5 T magnetic field.
• The presence of a 1.5 T magnetic field substantially modifies dose distributions at tissue interfaces.
• GPUMCD is suitable for dose modeling in MRI-Linac systems, but interface effects due to magnetic fields must be considered.