Date of Award
Master of Science
Jake Van Dyk
Radiobiological models are used to predict normal tissue complication probabilities (NTCP) in irradiated healthy tissues. The dependence of these models on the inherent uncertainties of dose-response data is not well established. This thesis has evaluated the impact of these uncertainties in the Lyman model for NTCP in lungs. Datasets were generated, based on previously published clinical and animal data, to simulate dose and volume uncertainties in numerous animal experiments and clinical trials of lung irradiations, and fit to the Lyman model to determine parameter estimates. Results from datasets with the same level of uncertainty were compared to the reference values, and to the results from simulations of other levels of uncertainty. The results clearly indicate that NTCP estimates should only be made in cases that have the same level of uncertainty and are within the same data range used to determine the parameter values that are part of NTCP calculations.
Kurjewicz, Laryssa M., "Radiobiological Modeling and the Impact of Treatment Uncertainties" (2009). Digitized Theses. 3860.