Matheurística para os problemas da geometria e da intensidade em IMRT
Radiotherapy is a form of treatment of cancerous tissues by means of ionizing radiation. The fundamental idea of radiotherapy treatment is to administer a dose of radiation to the tumor region sufficient to destroy it, sparing the anatomical healthy structures. The complete planning of a radiotherap...
Na minha lista:
| Autor principal: | |
|---|---|
| Outros Autores: | |
| Formato: | bachelorThesis |
| Idioma: | pt_BR |
| Publicado em: |
Universidade Federal do Rio Grande do Norte
|
| Assuntos: | |
| Endereço do item: | https://repositorio.ufrn.br/handle/123456789/34212 |
| Tags: |
Adicionar Tag
Sem tags, seja o primeiro a adicionar uma tag!
|
| Resumo: | Radiotherapy is a form of treatment of cancerous tissues by means of ionizing radiation. The fundamental idea of radiotherapy treatment is to administer a dose of radiation to the tumor region sufficient to destroy it, sparing the anatomical healthy structures. The complete planning of a radiotherapy treatment consists of the following steps: (I)
select the beam angles, (II) calculate the intensity of the beams, and (III) to define a radiation delivery sequence (ACOSTA et al., 2008). Such steps can be approached as NP-hard optimization problems, having different mathematical models and a variety of algorithms and techniques applicable for such. In general, these models propose objective
functions that somehow penalize excess of radiation in healthy and noble tissues and insufficient dose in the tumor. After literature review, the model presented in (OBAL, 2016) was adopted. Such model refers to problems (I) and (II), commonly called geometry problem and intensity problem, respectively. Its solution methods consists of hybridization
of metaheuristics with Simplex, an approach known in the literature as a matheuristic. The metaheuristics perform the search for beam sets, whereas the Simplex calculates the intensity of the beams, using weighting factors for the objective functions. Based on this work, this monograph proposes a matheuristic that hybridizes Tabu Search accompanied
by the ejection chain technique with the Simplex method. The two methods are employed in the same way as in (OBAL, 2016), but stands out the differentiation between searches. In the Tabu Search proposed here, besides the neighborhood exploration through the ejection chain, it was also decided by the exploration of different sizes of beam sets, as
well as the use of random-restart of the solution. For the evaluation of the proposed algorithm, adapted test cases from (BREEDVELD; HEIJMEN, 2017) are used. Rather than considering cases in their entirety, only a subset of regions is handled by the algorithm.The analysis of the results suggests that the approach proposed in this work is able to obtain radiotherapeutic treatments of better quality compared to the algorithm of (OBAL, 2016). |
|---|