Uncertainties in Monte Carlo-based absorbed dose calculations for an experimental benchmark

Area of Science:

• Medical Physics
• Radiation Dosimetry
• Computational Physics

Background:

• General-purpose Monte Carlo codes like EGSnrc are crucial for radiation therapy dosimetry.
• Existing experimental benchmarks often involve normalization, lacking absolute verification.
• There is a need for absolute benchmarks to rigorously assess Monte Carlo code accuracy.

Purpose of the Study:

• To perform an absolute benchmark experiment for verifying radiation transport calculations in radiotherapy.
• To validate the accuracy of the EGSnrc Monte Carlo code using experimental data.
• To analyze and quantify uncertainties in Monte Carlo simulations for dosimetry applications.

Main Methods:

• An absolute benchmark experiment using a thimble ionization chamber in a solid phantom irradiated by high-energy bremsstrahlung.
• Precise determination of accelerator and experimental setup characteristics.
• Monte Carlo simulation using EGSnrc, with detailed uncertainty analysis based on the Guide to the expression of uncertainty in measurement.

Main Results:

• Significant uncertainty contributions identified in radiation source energy, photon cross sections, and I-values.
• The combined standard uncertainty of the EGSnrc Monte Carlo calculation was estimated at 0.78%.
• Calculated absorbed dose closely matched the experimental result, with combined standard uncertainty <1%.

Conclusions:

• The accuracy of the EGSnrc Monte Carlo code for radiation therapy dosimetry is confirmed.
• The developed benchmark methodology can be applied to validate other Monte Carlo codes.
• Absolute verification is essential for ensuring the reliability of computational dosimetry in clinical practice.