Introduction: One crucial point when calculating the distribution of doses with ions is the uncertainty of the Bragg peak. The proton ranges in determined geometries like homogeneous phantoms and detector geometries can be calculated with a number of various parameterization models. Several different parameterizations of the range-energy relationship exist, with different levels of accuracy and complexity. For benchmarking purposes and calibration of proton range, it is consequential to have an accurate computation scheme between ranges and energies. In this setting, Monte Carlo simulations became important more and more in order to evaluating treatment plans and dose distributions. High-resolution energy-range tables are created using the PSTAR database. The aim of this study is to calculate proton range in the range of therapeutic energy in a cubic water phantom with a submillimeter accuracy.
Materials and Methods: Various Monte Carlo packages are available today that are specifically developed for handling radiation transport problems. GATE (version8) was used in this study to model the geometry and composition of a phantom. Geometries dictated to the toolkit were a cubic water phantom (40*40*40 cm3), as the target sitting on the xy-plane with the z-axis as its axis of symmetry. The primary particle source, emitting protons, were in the proximity of the phantom base on the z-axis. Mono energetic proton pencil beams (50, 100, 150, 200 MeV) hit the phantom. Several physics lists are defined in the GATE that we used FTFP_BERT. The simulations were carried out for 106 proton histories that yielded better than 1% statistical errors.
Results: In the current study, the results of the Bragg Peak Profile for the energy range of 5-
200 MeV has been obtained. The range-energy relation was obtained by fitting the FTFP_BERT physics data. So far, many similar studies have been done in this regard, such as a study by Bozkurt using the MCNPX code. However, we investigated the overall energy range used in proton therapy and obtained the fit model using a greater bunch of data.
Conclusion: By comparing the results obtained for each energy with NIST data, and with using Shapiro-Wilk statistical test, we did not see any significant difference. It was also found by calculating the percentage difference obtained with the CSDA data available in the NIST library, with the highest difference of 0.5%.