ANALYSIS OF THE BOWTIE FILTER EFFECT ON THE BEAM CHARACTERISTICS OF THE VARIAN OBI CBCT USING THE EGSNRC MONTE CARLO SIMULATION

• Main Author: Setyawan, Dedy
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/71801
• Institution: Institut Teknologi Bandung

The use of Cone Beam Computed Tomography (CBCT) in radiation therapy is increasing. Despite this, CBCT can increase the dose delivered not only to tumors but also to healthy tissue. Therefore, CBCT dosimetry must be performed accurately. Particle transport on Varian Medical Systems' On-Board Imager (OBI) CBCT can be simulated with Electron Gamma Shower by National Research Council (EGSnrc) Monte Carlo method. In this study, the CBCT system used Varian OBI CBCT with the tube model G242. Beam quality analysis was performed on two CBCT head models, model A (without using a bowtie filter) and model B (using a bowtie filter) to determine the effect of using a bowtie filter on the resulting beam characteristics. In this study, 100 million and 1 billion electrons with an X-ray tube voltage of 120 kVp were used. The result showed that the more particles and component modules, the longer it takes to complete the simulation. Based on the fluence vs position graph and the energy fluence vs position graph, the beam shape is a cone beam with a diameter of about 8 cm in model A and 8.9 cm in model B. Based on the spectral distribution graph, there are four characteristic X-rays spectra for model A (100 million histories) and five characteristic X-rays spectra for model A (1 billion histories). Meanwhile, a bowtie filter in model B increases the attenuation and reduction of low-energy photons, resulting in a minimal energy shift in the spectral distribution and energy fluence distribution graphs. The average energy obtained from the mean energy distribution graph for model A are 70.7 keV (100 million histories) and 70.0 keV (1 billion histories), while for model B, they are 73.1 keV (100 million histories) and 73.2 keV (1 billion histories). Based on the angular distribution graph, most of the particles in model A and model B simulated at 14°. Most parameter values in model A are always larger than model B, with data differences ranging from 47% to 52.5%. These results showed that the EGSnrc Monte Carlo simulation successfully simulates the VARIAN OBI CBCT and investigates the resulting beam characteristics. This method also successfully explains the effect of using a bowtie filter since the usage of a bowtie filter reduces the scattering of X-rays, makes the distribution of X-rays more homogeneous, and reduces the detected charged particles.