Computation of a voxelized antropomorphic phantom from Computer Tomography slices and 3D dose distribution calculation utilizing the MCNP5 Code

摘要

The purpose of this work is to obtain the voxelization of a series of tomography slices in order toprovide a voxelized human phantom throughout a MatLab algorithm, and the consequent simulation of theirradiation of such phantom with the photon beam generated in a Theratron 780? (MDS Nordion) ~(60)Coradiotherapy unit, using the Monte Carlo transport code MCNP (Monte Carlo N-Particle), version 5. The projectprovides dose mapping calculations inside the voxelized antropomorphic phantom. Prior works have validatedthe cobalt therapy model utilizing a simple heterogeneous water cube-shaped phantom [1]. The referencephantom model utilized in this work is the Zubal phantom [2], which consists of a group of pre-segmented CTslices of a human body.The CT slices are to be input into the Matlab program which computes the voxelizationby means of two-dimensional pixel and material identification on each slice, and three-dimensional interpolation,in order to depict the phantom geometry via small cubic cells. Each slice is divided in squares with the size ofthe desired voxelization, then the program searches for the pixel intensity with a predefined material at eachsquare, making a subsequent three-dimensional interpolation. At the end of this process, the program produces avoxelized phantom in which each voxel defines the mixture of different materials that compose it. The output ofthis code follows the MCNP input deck format and is integrated in a full input model including the ~(60)Coradiotherapy unit. Dose rates are calculated using the MCNP5 tool FMESH, superimposed mesh tally. Thisfeature allows to tally particles on an independent mesh over the problem geometry, and to obtain the lengthestimation of the particle flux, in units of particles/cm2·s (tally F4). Furthermore, the particle flux is transformedinto dose rate by using the conversion factors extracted from the NIST Physical Reference Data.