Arquitetura de indexação aplicada a servidores de PACS para processamento de imagens

• Ano de defesa: 2007
• Autor(a) principal: Cotrim, Daniel da Silva
• Orientador(a): Bezerra, Eduardo Augusto
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Pontifícia Universidade Católica do Rio Grande do Sul; Porto Alegre
• Palavras-chave em Português: INFORMÁTICA; PROCESSAMENTO DE IMAGENS - TÉCNICAS DIGITAIS; DIAGNÓSTICO POR IMAGEM; INFORMÁTICA MÉDICA; ARMAZENAMENTO DE IMAGENS; RECONHECIMENTO DE PADRÕES
• Link de acesso: http://hdl.handle.net/10923/1608

Several techniques for image processing have been developed for diagnosis aid support. However, in some cases these techniques are not applied to the images stored in PACS servers (Picture Archiving and Communication Systems). This scene establishes the viability to build a new storage and recovery indexing architecture having the capability of identifing the need for executing certain types of imaging processing for examinations of patients with features defined by user. It can be explained as a result of the intrinsic difficulty related to the integration of these techniques to the hospital production environment. A possible solution for this problem would be a virtual environment to make different processing features available to be executed directly in the PACS. This environment would have as goal to stimulate the medical support application. The proposed indexing architecture uses information from images stored in PACS, relating them to the pre-defined processing class, establishing an execution context. The information associated to each processing class is based on ontologies representing the formalization of the knowledge applied to each group of images or study. A case study has been developed in a PACS.