Análise remota do eletrocardiograma para detecção de eventos isquêmicos

• Ano de defesa: 2019
• Autor(a) principal: Bentes, Paulo César Lucena
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal do Rio de Janeiro; Brasil; Instituto Alberto Luiz Coimbra de Pós-Graduação e Pesquisa de Engenharia; Programa de Pós-Graduação em Engenharia Biomédica; UFRJ
• Palavras-chave em Português: Computação remota; Eventos isquêmicos; Exames eletrocardiográficos; CNPQ::ENGENHARIAS
• Link de acesso: http://hdl.handle.net/11422/13214

The evolution of technologies for remote services delivery over the Internet unveils a new frontier in the development of the knowledge needed to implement health prevention measures. In this study, a computational tool was conceived for the remote analysis of multiple lead electrocardiograms. As a proof of concept, a method for detecting ST-T segment changes related to ischemic episodes in remote computing is proposed. The architecture combines only open source software that allows incremental object-oriented programming and support multiuser services via the Web, focusing on system evolution within the academic world. The technique used to detect ischemic events favored low computational cost and storage of both data and metadata in a database. It was anchored in a method of interpolation by weighted least squares and histograms, capable of detecting the positions of the QRS complexes, and the respective positions of J points and T waves. These points were used as borderline positions in obtaining representative under curve areas for the subsequent detection of ischemic events in the leads present in the research file. After assessment with engineering students, we conclude that the platform, architecture, and programming techniques provide a satisfactory tool for ischemic event management that can be used to develop new biomedical signal processing techniques that support the risk assessment of myocardial dysfunction.