Desenvolvimento de um Sistema de Reconhecimento de Atividades Humanas e Monitoramento Remoto Utilizando um Dispositivo Vestível

• Ano de defesa: 2017
• Autor(a) principal: Coelho, Yves Luduvico
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal do Espírito Santo; BR; Mestrado em Engenharia Elétrica; Centro Tecnológico; UFES; Programa de Pós-Graduação em Engenharia Elétrica
• Palavras-chave em Português: Embedded Systems; Wearable Devices; Pattern Recognition; Human Activity Recognition; Remote Monitoring; Reconhecimento de atividades humanas; Monitoramento Remoto; Dispositivos Vestíveis; Sistemas embarcados (Computadores); Sensoriamento remoto; Reconhecimento de padrões; Eletrônica Industrial, Sistemas e Controles Eletrônicos; 621.3
• Link de acesso: http://repositorio.ufes.br/handle/10/9680

Factors such as the aging population and the consequent increase in the number of people with chronic diseases lead toan exponential increase of healthcarecosts, since the healthcaresystem must be able to serve an increasing number of people while maintaining the quality of the attendance. In order to reduce costsand improve quality, it is important tomove towardsa patient-centered healthcaresystem, in which it is possible todetect early warning signs, avoiding hospitalizations, as well as follow the patientsremotely, avoiding a stay in the hospital. In this context, remote monitoring devices become essential for gathering of important patient information and for making them available to the healthcare provider. The technological advancement regarding the miniaturization of sensors and the new low-power wireless communication technologiesencouragethe development of remote health monitoringsystems. Thiswork proposes the development of a system of human activity recognition andremote monitoring in three different approaches. For the first approach, an accuracy of 89.11%and a precision of 91.45%wereobtainedwhen classifyingsix different activities. For thelast two approaches, a complete structure of remote monitoringwas developed to monitor the user’s activity intensity, from thedata collection, in order to transfer it by e-mail to the health provider. Results demonstrate the efficacy of this system for human activity recognition and remote monitoring.