Uma técnica otimizada de clusterização para segmentação de imagens de TC de tórax de alta-resolução

• Ano de defesa: 2016
• Autor(a) principal: Porto, Marcelo Arrais
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Santa Maria; Brasil; Ciência da Computação; UFSM; Programa de Pós-Graduação em Informática; Centro de Tecnologia
• Palavras-chave em Português: Segmentação de imagens; Diagnóstico de imagens médicas; Tomografia computadorizada; Image segmentation; Medical images diagnosis; Computed tomography; CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: http://repositorio.ufsm.br/handle/1/12169

Lung segmentation is a fundamental step in many image analysis applications for lung diseases and abnormalities in thoracic computed tomography (CT). However, due to the large variations in pathology that may be present in thoracic CT images, it is difficult to extract the lung regions accurately, especially when the lung parenchyma contains extensive lung diseases. A major insight to deal with this problem is the existence of new approaches to cope with quality and performance. This paper presents an optimized superpixel clustering approach for high-resolution chest CT segmentation. The proposed algorithm is compared against some open source superpixel algorithms while a performance evaluation is carried out in terms of boundary recall and undersegmentation error metrics. The over-segmentation results on a Computed Tomography Emphysema Database demonstrates that our approach shows better performance than other three state-of-the-art superpixel methods.