Determining optimization of the initial parameters in Monte Carlo simulation for linear accelerator radiotherapy

Monte Carlo (MC) method is a well known calculation algorithm which can accurately assess the dose distribution for radiotherapy. The present study investigated all the possible regions of the depth-dose or lateral profiles which may affect the fitting of the initial parameters (mean energy and the radial intensity (full width at half maximum, FWHM) of the incident electron). EGSnrc-based BEAMnrc codes were used to generate the phase space files (SSD=100 cm, FS=40×40 cm2) for the linac (linear accelerator, Varian 21EX, 6 MV photon mode) and EGSnrc-based DOSXYZnrc code was used to calculate the dose in the region of interest. Interpolation of depth dose curves of pre-set energies was proposed as a preliminary step for optimal energy fit. A good approach for determination of the optimal mean energy is the difference comparison of the PDD curves excluding buildup region, and using D(10) as a normalization method. For FWHM fitting, due to electron disequilibrium and the larger statistical uncertainty, using horn or/and penumbra regions will give inconsistent outcomes at various depths. Difference comparisons should be performed in the flat regions of the off-axis dose profiles at various depths to optimize the FWHM parameter.

► Crucial parameters were investigated using a logical system procedure.
► Interpolation of depth dose curves using pre-set energies as a first step.
► Comparisons using PDD curves normalized by D(10), excluding buildup region.
► FWHM is sensitive to horn/penumbra regions, but they give inconsistent values.
► Optimizing the FWHM requires the flat regions at various depths.