Monte Carlo Dose Verification of an X-Ray Beam in a Virtual Water Phantom

Monte Carlo (MC) methods are widely accepted as the most accurate technique for calculating dose distributions in radiation therapy physics. Simulating the particle transport through the treatment head of a linear accelerator utilizing a MC based code is both a widespread and practical approach to determining detailed clinical beam characteristics such as the energy, angular and spatial distribution of particles which are needed to properly quantify dose. One particular and versatile MC code, the Monte Carlo N-Particle (MCNP) radiation transport code, developed by Los Alamos National Laboratory, has been commonly used to model ionizing radiations for medical physics applications. \n\tIn this thesis, a Varian 2100C linear accelerator (linac) is modeled and the electron and photon transport through the primary components of the treatment head are simulated using MCNP Version 5_1.3. The 6 MV photon spectra was characterized in a standard 10 x 10 cm2 field and subsequent dose calculations were made in a Virtual Water (VW) phantom. Energy fluence, percent depth dose and beam profile measurements were taken in a modeled VW phantom and the calculated data was compared to measured reference data. In addition, a human phantom was modeled for future dose calculations using the modeled linac. \n\tThe linac model created can incorporate different beam energies for determining the dose distribution of multiple beam treatments in phantoms for standard 6 MV plans. The adaptability of this MCNP model allows for any number of geometries and sources encountered in medical physics to be computed and applied with relative ease. Future studies can involve adding complex multi-leaf collimator beam shaping and calculating the dose in human phantom models, which would serve as a basis for studies involving MCNP modeling for dose optimization in medical physics applications.