- 71 views
- 155 downloads
Automatic Optimization of Treatment Dosimetry to Improve Visual Outcomes in Episcleral Plaque Brachytherapy
-
- Author / Creator
- Han, Gawon
-
Ocular plaque brachytherapy is an effective treatment option for medium-sized ocular melanomas, showing equivalent survival relative to enucleation, while providing added benefits of eye preservation and possible vision retention. Currently the treatment planning system for plaque brachytherapy, known as Plaque Simulator, generates treatment plans using uniform seed strengths which requires the planner to modify the plan by manually adjusting the seed strengths and locations to reduce doses to the critical structures while maintaining the desired tumor coverage. This demands planning expertise, can be time-consuming, and may not always provide the best possible plan. To overcome these challenges, this thesis applies an automated dose optimization algorithm, known as simulated annealing (SA) algorithm, to inverse plan plaque brachytherapy treatments. Firstly, the SA algorithm was verified by solving two problems: 1) dose optimization using uniform loading, and 2) maximizing dose uniformity across tumor base with differential loading. The first problem allowed to verify the correct implementation of the tumor and seed geometry, dose calculation, and the general steps of the SA algorithm. Using the SA algorithm, uniform seed strengths to deliver the prescription dose to tumor apex were obtained and were manually input into Plaque Simulator to compare the resulting dose to apex for various tumor and plaque sizes. The difference in dose to apex between the two systems were found to agree within 4.5% for all scenarios and arises due to the differences in the TG-43 dosimetry parameters and seed coordinates used in both systems. The second problem evaluated the potential of the algorithm to find minima in the energy function using differential seed strengths. The algorithm from the first problem was modified to find differential seed strengths that can maximize dose uniformity across the tumor base. Averaging over different plaque sizes, the reduction in basal dose variation for 3.5, 5, and 8 mm heights from using uniform strengths, were found to be 33.1%, 33.3%, and 27.1%, respectively. Lastly, the problem of finding differential seed strengths to reduce doses to the critical structures while maintaining desired tumor coverage was investigated. The SA algorithm was used to reduce dose to a single point of interest (POI), and then two POIs representative of critical structures; the algorithm was found to be able to reduce doses, to a degree that depends on the tumor and plaque size used. The algorithm was then further developed to handle shifted plaques and elliptical tumor base shapes in its modelling, which have been verified by benchmarking each feature as well as the two together in optimization. Moreover, by applying the algorithm to numerous clinical scenarios, it was verified that implementing the developed optimization routine into clinical cases is practical.
-
- Graduation date
- Fall 2019
-
- Type of Item
- Thesis
-
- Degree
- Master of Science
-
- License
- Permission is hereby granted to the University of Alberta Libraries to reproduce single copies of this thesis and to lend or sell such copies for private, scholarly or scientific research purposes only. Where the thesis is converted to, or otherwise made available in digital form, the University of Alberta will advise potential users of the thesis of these terms. The author reserves all other publication and other rights in association with the copyright in the thesis and, except as herein before provided, neither the thesis nor any substantial portion thereof may be printed or otherwise reproduced in any material form whatsoever without the author's prior written permission.