Comparison of the different physics lists in the Gate code in calculating proton beam range by comparing with NIST data

Introduction: When calculating dose distributions with ion beams, a critical point is the uncertainty of the Bragg-peak range. Recently great effort is devoted to enhance the accuracy of the computation for proton treatments without increasing the calculation time. The highest accuracy in these calculations is obtained by Monte Carlo methods. Methods: In this study, GATE code was used to compute absorbed dose of protons in the energy range of 30-250 MeV and with energy step of 10 MeV. A water phantom was modeled in a vacuum world volume. Currently, Several physics lists have been defined in the GATE code which the user must select the appropriate model as needed. In this study, we examined the accuracy of different physics list: QGSP_BIC_EMY, QGSP_BERT_HP_EMY, QBBC, FTFP_BERT, and QGSP_BIC_EMZ and the results were compared with the NIST data. The simulations were carried out for 10^6 proton histories that yielded better than 0.3% statistical errors.Results:In this study, in order to compare the observed differences between the results obtained from each of the physics lists with the NIST library data we first used SPSS software. Using the Mann-Whitney test, the p-value was equal to 0.831, indicating that there is no significant difference between the values obtained from each physics with the NIST library data. However, since the results have to be evaluated with high precision, other analytical methods have also been used. In all of the studied energy ranges, in terms of comparing the difference between the results of the proton range and the statistical simulation uncertainty, QBBC, FTFP_BERT, and QGSP_BIC_EMZ physics lists showed a greater deviation than the NIST data. In this study, we used the least squares method in order to verify more precisely. With this method, the QGSP_BIC_EMY showed the minimum difference with the NIST data. Therefore, our suggested physics for proton therapy applications is QGSP_BIC_EMY.Conclusion: The results were found to be ±0.1% compared to the data from the NIST. It is safe to conclude that this approach can be extended to determine dosimetric quantities for other media, energies and charged particle types.

Keywords:

Gate code , Monte Carlo simulation , proton therapy , physics list , validation