Year

2007

Degree Name

Doctor of Philosophy

Department

School of Engineering Physics - Faculty of Engineering

Abstract

The objectives of this study were to improve our understanding of the contamination arising in conventional radiotherapy treatment from the various principal processes involved. The evolution of the contamination was investigated in two main ways: i) experimental measurements; and ii) the Monte Carlo method. The magnetic field strengths in this improved design were intended to result in more uniform magnetic flux densities in the area of interest, with the prediction of a greater volume where the electron contamination was effectively removed by our magnetic deflector device. The magnetic field strengths obtained by the magnetic deflector will theoretically give rise to electron deflection radii that should cause the majority of electron contamination to exit the treatment field. An enhancement of the electron dose was never experimentally observed in the irradiated area, and a percentage reduction of the skin and subcutaneous dose up to 34% with the NdFeB magnetic device was seen for a 20 x 20 cm2 field size. The elimination of significant electron doses due to contaminant electrons down to a depth of a few millimetres was obtained with this newly designed magnetic deflector device. In the study, the experiments were verified by an Attix chamber and radiographic film. The surface dose was increased as the field size was increased in an open field and when a Perspex tray was placed in the beam, with the increase especially significant in the case where there was both a Perspex tray and a larger field size. The Perspex tray or a wedge filter eliminate secondary electrons and generate new electrons at the same time, however, when combined with magnetic field the surface dose is reduced significantly. Results are also shown for the surface dose profile in two dimensions (x and y-axis) with the surface dose showing a decrease at all sites within the treatment field due to the magnetic deflector device, not only for an open field, but also when a wedge or a Perspex tray is in the beam. Calculation and analysis of spectra of deflected electrons in photon beams from the linear accelerator treatment head were investigated. Calculating such spectra with more accuracy requires knowledge of the characteristics of the electron beam incident on the target as well as better equipment for modelling the linear accelerator. We used the Monte Carlo method performed with BEAMnrc and DOSXYZnrc code to derive estimates for the average energy deposited in the system. Monte Carlo modelling of photon beams was achieved and adjusted for two parameters: AE = ECUT = 0.521 MeV and AE = ECUT = 0.700 MeV by matching the Monte Carlo calculated depth dose and beam profile data with the measured data. The capability of the Monte Carlo program in evaluating dose distribution has been verified by comparison with measurements in a water phantom and with radiographic film. The comparisons were performed for percentage of the build-up dose for various field sizes. Ionisation measurements were made in a solid water phantom by means of an Attix chamber for experiments to determine the dose in the build-up region. The measurement of skin dose uses an Attix parallel plate ionisation chamber, which is primarily used as the benchmark chamber in solid-water phantom dose build-up measurements. Monte Carlo simulations were performed to generate data to predict the dose distribution for 6 MV x-rays. Investigation of dose components of electron spectra are compared between calculated and measured dose distributions. From the Monte Carlo calculations and measurements on the surface and in the build-up region for 6 MV x-ray beams based on our results, we conclude that our optimised simulation model represents the beam emerging from the treatment head and the calculated percentage depth doses in such a way that there is a satisfactory match with the experimental measurements for the same irradiation set-ups.

Download

02Whole.pdf (6139 kB)

Share

COinS
 

Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.