Energy deposition and dose enhancement using Monte Carlo derivative sampling: applications in brachytherapy

Abstract

Energy deposition and radiation dose distribution by the use of gold, a high-Z biocompatible element in water solution, is estimated as a function of source energy typical of brachytherapy sources (15 keV, 20 keV, 30 keV, 40 keV, 80 keV, 90 keV, 150 keV, 300 keV, 1 MeV), solution concentration (5-25 mg Au/g H 2 O), and solution placement (1-2 cm concentric shells). Monte Carlo (MC) simulations are carried out with the MCNP5 code, compared with other widely used MC codes such as PENELOPE and GEANT, to validate the dose estimates, which may vary considerably due to artifacts and data libraries, and extended to a sensitivity analysis using perturbation estimates. The energy deposition and radiation dose are estimated as a function of monoenergetic source radiation energy, and concentration of gold in solution. MC simulation is carried out in the coupled photon-electron radiation mode for x-rays emanating from a radiation source implanted in a water cell, for which results are valid for a cancer cell modelled by a spherical water phantom. For carrying out sensitivity studies, the Monte Carlo perturbation feature with material perturbations was used to sample derivatives in a single run which were used in a Taylor series to estimate both dose and Dose Enhancement Factor (DEF) from single MC runs. Close agreement was found between dose estimates from MCNP5, PENELOPE and GEANT, in spite of artifacts such as cut-offs in electron transport. It was also found that dose increases with energy of a source, and that dose enhancement, for a given concentration, decreases with source energy. The perturbation estimates result in enhanced computational efficiency.

Relative dose-response from solid-state and gel dosimeters through Monte Carlo simulations

“…Thus, computational simulations can be performed as an alternative to experimental methods to obtain valuable information with the prediction of computational results [31][32][33][34][35]. Several Monte Carlobased codes are used in physics research [36][37][38][39][40][41]. The MCNPX code was developed to study particle physics and nuclear reactions, so they simulate an extensive set of particles in a broad energy spectrum [42][43][44][45] and many medical physics applications [46][47][48][49][50].…”