Avaliação da curvatura escoliótica utilizando reconhecimento de padrões elípticos em imagens de raio X

• Ano de defesa: 2018
• Autor(a) principal: Coelho, Júlio Cézar
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; Brasil; Programa de Pós-graduação em Engenharia Elétrica
• Palavras-chave em Português: Encaixe de elipse; Algoritmo genético; Processamento digital de imagem; Escoliose; Reconhecimento de padrões; Vértebras; Ellipse fitting; Genetic algorithm; Vertebrae; Scoliosis; Pattern recognition; Digital image processing; Engenharia elétrica; Elipse (Geometria); Processamento de imagens - Técnicas digitais; CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO::METODOLOGIA E TECNICAS DA COMPUTACAO::PROCESSAMENTO GRAFICO (GRAPHICS)
• Link de acesso: https://repositorio.ufu.br/handle/123456789/23099; http://dx.doi.org/10.14393/ufu.te.2018.796

Scoliosis is a disease characterized by a deformation of the spine. This deformation is composed of an abnormal lateral bending and a rotation of some vertebrae. The method more used for assessment of quantity of scoliosis is the angle of Cobb. Many ways were searched for assessment this deformity, as angles, distances, area, circles rays and, finally, ellipses. This paper proposes a methodology able of comparing the deformation of the spine with an ellipse, thus providing sufficient mathematical and visual parameters for an auxiliary view on the degree of curvature and its progression. The results were possible using genetic algorithm able to test many ellipses possible for the perfect fitting, having mathematical and visual parameters that are compound by characteristic parameters of ellipse that produces an index that gives a general view of the deformity. The results of this study indicates that the proposed technique is able to recognize with accuracy elliptical patterns in different types spine with scoliosis from x-ray images.