Monte Carlo Comparison of Dose-to-Air-Kerma Ratios for Kilovoltage Ionization Chamber Dosimetry
Abstract
Summary
Background: Comparative studies show that at kilovoltage energies, EGSnrc and Geant4 can differ by several percent in ionization chamber response, particularly in small air cavities and steep dose gradients. These discrepancies are primarily attributed to differences in electron transport and atomic relaxation modeling rather than photon cross sections.
Purpose
The aim of this study was to compare dose-to ati-kerma ratios in kilovoltage beams to assess differences between EGSnrc and Geant4 Monte Carlo codes.
Methods
Spectra for kilovoltage beams with energies ranging from 15 to 50 kVp (TW15-TW50) were taken from PTW-Freiburg laboratories and used as input in the simulations. The PTW 34013 ionization chamber was modeled with E-Brachy, a Geant4-based software package, to accurately model the setup. The detector cavity was positioned 30 cm from a point source that created a cone-shaped beam of 3 cm in diameter on the chamber. Dose to the detector cavity Dgas was scored as well as the air kerma at the same point without the detector (Kair). The ratio of Kair) to Dgas was calculated. The uncertainty in the air cavity dimensions inside the chamber was considered. The results were then compared with simulations previously carried out using EGSnrc unders the same conditions.
Results
Values of Dgas /Kair for the calibration beams, ranging from 77% to 93%, which were systematically lower than the corresponding values computed by the EGSnrc package by 5-10%. However, additional simulation histories are needed to reduce uncertainty below the 1% level. The MC-calculated TG-43 depth-dose distributions agreed with measurements within 5%.
Conclusion
This work underscores the importance of rigorous Monte Carlo validation, including cross-code comparison and the reduction of statistical uncertainty, before such an approach can be disseminated for clinical use.