Faculty of Graduate Studies and Research in partial fblfillment of the requirements for the degree of L'auteur conserve la propriété du droit d'auteur qui protège cette thèse. Ni la thèse ni des extraits substantiels de celle-ci ne doivent être imprimés ou autrement reproduits sans son autorisation. Abstract The head scatter factor, S, is an important dosimetric quantity used in radiation therapy dose calculation. It is empirically determined and its field size dependence reflects changes in

more »

... ton scatter fkom components in the linac treatrnent head. In this work a detailed study of the physical factors influencing the determination of S, was performed with particular attention given to asymmetric field geometries. Ionization measurements for 6 and 18 MV photon bearns were made to examine the factors which determine S,. These include: phantom size and material, collimator backscatter, non-lateral electronic equilibriurn (LEE) conditions, electron contamination, collimator-exchange, photon energy, flattening filter and off-axis distance (OAD). Results indicated that LEE is not required for S, measurements if electron contamination is minimized. Brass caps or polystyrene miniphantoms can both be used in S" measurements provided the phantom thickness is large enough to stop contaminant electrons. Backscatter radiation effects into the monitor chamber were found to be negligible for the Siemens linac. It was found that the presence and shape of the flattening filter had a significant effect on the empirically determined value of S, was also s h o w to be a function of OAD, particularly for small fields. For fields larger than 12 x 12 cm2 S, was independent of OAD. A flattening filter m a s mode1 was introduced to explain qualitatively the above results. A detailed Monte Carlo simulation of the Siemens KD2 linac head in 6 MV mode was perfonned to investigate the sources of head scatter which contribute to the measured S,. The simulated head components include the flattening filter, the electron beam stopper, the primary collirnator, the photon monitor chamber and the secondary collimators. The simulations showed that the scatter from the head of the Siemens linac is a cornplex hinction of the head components. On the central avis the flattening filter played the dominant role in the contributing to scatter. However this role was significantly reduced off-axis and other head components, such as the electron beam stopper and the primary collimator, became more important. The role of the mirror and ion chamber was relatively rninor. Scatter from the secondaxy collimators was shown to be a function of the filed size and the position of the collimators in the treatment head. They were also found to play a dual role, both as a scatter source and as an attenuator for scatter produced upstream in the linac head. A closed form model, based on the work of Yu and Slobada, was developed to estimate head scatter factors for on-and off-axis asymmetnc fields. The model requires three parameten to fit the measured data. The fint, a constant c, has a physical significance and is independent of energy and off-axis distance. The second, g, shows a smdl variation with the energy and OAD while the third parameter, the primary-to-scatter ratio, is strongly dependent on energy and off-axis distance. Cornparison of S" predicted by the model, to measurement for a large range of symmetric and asymmetric fields showed excellent agreement. A maximum of 0.7% discrepancy was observed at 12 cm OAD.