Otimização aplicada ao planejamento de dosagem radioterápica para o tratamento do câncer

• Ano de defesa: 2019
• Autor(a) principal: Vivas, Angelica Maria Narvaez [UNIFESP]
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de São Paulo (UNIFESP)
• Palavras-chave em Português: Nonlinear Optimization; Radiotherapic Dosing Planning; Intensity Modulated Radiation Therapy; Beam Angle Optimization; Otimização Não Linear; Planejamento De Dosagem Radioterápica; Radioterapia De Intensidade Modulada; Otimização Dos Ângulos De Feixe
• Link de acesso: https://sucupira.capes.gov.br/sucupira/public/consultas/coleta/trabalhoConclusao/viewTrabalhoConclusao.jsf?popup=true&id_trabalho=8320245; https://repositorio.unifesp.br/handle/11600/59448

When a person is diagnosed with cancer and receives the prescription of treatment with radiotherapy, it is necessary to draw up a treatment plan. For this, the patient undergoes a 3D computed tomography scan, in order to delineate the organs affected by the tumor. Based on tomographic imaging, the oncologist is able to predict the distribution of radiation doses throughout the patient’s body. In the late 1990s, a radiotherapy technique called IMRT (Intensity Modulated Radiation Therapy) was developed, which aims to concentrate high doses of radiation only in the tumor regions, sparing as much as possible the adjacent healthy organs. The IMRT is based on linear accelerators and uses multiple beams of angular radiation and non-uniform intensities along with the use of mathematical optimization methods to determine the optimal distribution of radiation doses. Among the several variants of the IMRT technique, we adopted, in this work, the BAO (Beam Angle Optimization) variant. In the BAO, given an initial set of candidate radiation beams, the objective is to determine which of them are optimal, that is, those beams that will have a significant contribution to the treatment planning. To solve the associated optimization problem, we use a method called Adaptive ℓ2, 1-Minimization. We will make a comparative analysis between some treatment plans for prostate cancer obtained with this method, starting from several different sets of candidate radiation beams.