{"id":{"repo_id":"rice","oai_identifier":"oai:repository.rice.edu:1911/118641"},"canonical_url":"https://search.dev.ndltd.org/etd/rice/oai:repository.rice.edu:1911/118641","repository":{"repo_id":"rice","name":"Rice University","base_url":"https://repository.rice.edu/server/oai/request"},"display":{"title":"Temporally Feathered Radiation Therapy under Uncertainty","abstract":"This thesis focuses on radiation therapy planning through novel stochastic optimization models that account for biological heterogeneity and uncertainty in organ-at-risk (OAR) responses. Building on the temporally feathered radiation therapy (TFRT) strategy, we develop a personalized, biologically informed treatment framework that dynamically adjusts dose intensities and rest periods based on tissue-specific recovery potential. The model incorporates inter-organ variability using the linear-quadratic model within a dynamic normal tissue complication probability framework, allowing for tailored protection of sensitive tissues without compromising tumor coverage. To address uncertainty in patient-specific parameters, we propose two stochastic models: a risk-neutral formulation that maximizes expected OAR recovery and a risk-averse approach minimizing worst-case toxicity using Conditional Value-at-Risk. A modified L-shaped algorithm is developed to solve the resulting nonconvex problems more efficiently than conventional solvers. Clinically, these models outperform standard intensity modulated radiation therapy, with the risk-averse version protecting critical structures such as the spinal cord and brainstem, and the risk-neutral version improving outcomes for the parotid and structures associated with speech and swallowing. Because stochastic TFRT models can be computationally demanding as the number of scenarios increases, we introduce scenario generation techniques based on Wasserstein and fused Gromov-Wasserstein distances to approximate complex stochastic processes effectively. To solve the resulting models, we design block coordinate descent algorithms and demonstrate their performance across applications in stochastic TFRT, portfolio management, and capacity expansion.","abstract_html":"This thesis focuses on radiation therapy planning through novel stochastic optimization models that account for biological heterogeneity and uncertainty in organ-at-risk (OAR) responses. Building on the temporally feathered radiation therapy (TFRT) strategy, we develop a personalized, biologically informed treatment framework that dynamically adjusts dose intensities and rest periods based on tissue-specific recovery potential. The model incorporates inter-organ variability using the linear-quadratic model within a dynamic normal tissue complication probability framework, allowing for tailored protection of sensitive tissues without compromising tumor coverage. To address uncertainty in patient-specific parameters, we propose two stochastic models: a risk-neutral formulation that maximizes expected OAR recovery and a risk-averse approach minimizing worst-case toxicity using Conditional Value-at-Risk. A modified L-shaped algorithm is developed to solve the resulting nonconvex problems more efficiently than conventional solvers. Clinically, these models outperform standard intensity modulated radiation therapy, with the risk-averse version protecting critical structures such as the spinal cord and brainstem, and the risk-neutral version improving outcomes for the parotid and structures associated with speech and swallowing. Because stochastic TFRT models can be computationally demanding as the number of scenarios increases, we introduce scenario generation techniques based on Wasserstein and fused Gromov-Wasserstein distances to approximate complex stochastic processes effectively. To solve the resulting models, we design block coordinate descent algorithms and demonstrate their performance across applications in stochastic TFRT, portfolio management, and capacity expansion.","abstract_has_math":false,"creators":["Karagoz, Aysenur"],"institution":"Rice University","degree_name":"Doctor of Philosophy","degree_level":"Doctoral","degree_discipline":"Engineering","degree_department":null,"school":null,"contributors":[],"advisors":["Schaefer, Andrew J"],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025-07-18","date_published":"2025-07-18","updated_at":"2026-07-24T04:10:34Z","subjects":["cancer treatment optimization","scenario generation","optimization under uncertainty"],"languages":["eng"],"rights":["Copyright is held by the author, unless otherwise indicated. Permission to reuse, publish, or reproduce the work beyond the bounds of fair use or other exemptions to copyright law must be obtained from the copyright holder."],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/1911/118641","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Schaefer, Andrew J"]},{"key":"dc:creator","label":"Author","values":["Karagoz, Aysenur"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2025-09-03T21:26:53Z"]},{"key":"dc:date.issued","label":"Date","values":["2025-07-18"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Doctoral"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Doctor of Philosophy"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["Rice University"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["cancer treatment optimization","scenario generation","optimization under uncertainty"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["eng"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright is held by the author, unless otherwise indicated. Permission to reuse, publish, or reproduce the work beyond the bounds of fair use or other exemptions to copyright law must be obtained from the copyright holder."]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/1911/118641"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["This thesis focuses on radiation therapy planning through novel stochastic optimization models that account for biological heterogeneity and uncertainty in organ-at-risk (OAR) responses. Building on the temporally feathered radiation therapy (TFRT) strategy, we develop a personalized, biologically informed treatment framework that dynamically adjusts dose intensities and rest periods based on tissue-specific recovery potential. The model incorporates inter-organ variability using the linear-quadratic model within a dynamic normal tissue complication probability framework, allowing for tailored protection of sensitive tissues without compromising tumor coverage. To address uncertainty in patient-specific parameters, we propose two stochastic models: a risk-neutral formulation that maximizes expected OAR recovery and a risk-averse approach minimizing worst-case toxicity using Conditional Value-at-Risk. A modified L-shaped algorithm is developed to solve the resulting nonconvex problems more efficiently than conventional solvers. Clinically, these models outperform standard intensity modulated radiation therapy, with the risk-averse version protecting critical structures such as the spinal cord and brainstem, and the risk-neutral version improving outcomes for the parotid and structures associated with speech and swallowing. Because stochastic TFRT models can be computationally demanding as the number of scenarios increases, we introduce scenario generation techniques based on Wasserstein and fused Gromov-Wasserstein distances to approximate complex stochastic processes effectively. To solve the resulting models, we design block coordinate descent algorithms and demonstrate their performance across applications in stochastic TFRT, portfolio management, and capacity expansion."]},{"key":"dc:format.mimetype","label":"Dc Format Mimetype","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Temporally Feathered Radiation Therapy under Uncertainty"]}]}],"canonical_facts":{"dc:contributor.advisor":["Schaefer, Andrew J"],"dc:creator":["Karagoz, Aysenur"],"dc:date.accessioned":["2025-09-03T21:26:53Z"],"dc:date.issued":["2025-07-18"],"dc:description.abstract":["This thesis focuses on radiation therapy planning through novel stochastic optimization models that account for biological heterogeneity and uncertainty in organ-at-risk (OAR) responses. Building on the temporally feathered radiation therapy (TFRT) strategy, we develop a personalized, biologically informed treatment framework that dynamically adjusts dose intensities and rest periods based on tissue-specific recovery potential. The model incorporates inter-organ variability using the linear-quadratic model within a dynamic normal tissue complication probability framework, allowing for tailored protection of sensitive tissues without compromising tumor coverage. To address uncertainty in patient-specific parameters, we propose two stochastic models: a risk-neutral formulation that maximizes expected OAR recovery and a risk-averse approach minimizing worst-case toxicity using Conditional Value-at-Risk. A modified L-shaped algorithm is developed to solve the resulting nonconvex problems more efficiently than conventional solvers. Clinically, these models outperform standard intensity modulated radiation therapy, with the risk-averse version protecting critical structures such as the spinal cord and brainstem, and the risk-neutral version improving outcomes for the parotid and structures associated with speech and swallowing. Because stochastic TFRT models can be computationally demanding as the number of scenarios increases, we introduce scenario generation techniques based on Wasserstein and fused Gromov-Wasserstein distances to approximate complex stochastic processes effectively. To solve the resulting models, we design block coordinate descent algorithms and demonstrate their performance across applications in stochastic TFRT, portfolio management, and capacity expansion."],"dc:format.mimetype":["application/pdf"],"dc:identifier.uri":["https://hdl.handle.net/1911/118641"],"dc:language.iso":["eng"],"dc:rights":["Copyright is held by the author, unless otherwise indicated. Permission to reuse, publish, or reproduce the work beyond the bounds of fair use or other exemptions to copyright law must be obtained from the copyright holder."],"dc:subject":["cancer treatment optimization","scenario generation","optimization under uncertainty"],"dc:title":["Temporally Feathered Radiation Therapy under Uncertainty"],"dc:type":["Thesis"],"thesis:degree_discipline":["Engineering"],"thesis:degree_level":["Doctoral"],"thesis:degree_name":["Doctor of Philosophy"],"thesis:institution_name":["Rice University"]},"updated_at":"2026-07-24T04:10:34Z"}