Utilização da análise discriminante linear para criar um modelo linear de envelhecimento na execução de tarefas motoras por meio de sensores inerciais posicionados na mão e no antebraço

• Ano de defesa: 2023
• Autor(a) principal: Gonçalves, Veronica Lima
• 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: Envelhecimento humano; Sensores inerciais; Análise discriminante linear; Human aging; Inertial sensors; Linear discriminant analysis; CNPQ::ENGENHARIAS::ENGENHARIA BIOMEDICA; ODS::ODS 3. Saúde e bem-estar - Assegurar uma vida saudável e promover o bem-estar para todos, em todas as idades.
• Link de acesso: https://repositorio.ufu.br/handle/123456789/41046; https://doi.org/10.14393/ufu.te.2023.8079

One of the characteristics of the aging of the population is the loss of muscle and bone mass, which worsens the possibility of the emergence of neuromuscular pathologies. In order to help understand neuromuscular changes with aging, motor skills are characterized using inertial sensors, accelerometers, gyroscopes, magnetometers and electromyography. In this study, the objective was to verify a significant difference between the groups of young and elderly people through manual movements and whether the combination of characteristics can produce a linear correlation with the different age groups. Method: 99 individuals participated in the research, divided into 8 groups and grouped by age group. Data collection was carried out using inertial sensors positioned on the back of the hand and the back of the forearm). Initially the participants were divided into groups of young and old to check whether the characteristics could discriminate the groups, later the characteristics were combined using Linear Discriminant Analysis (LDA), giving rise to a single characteristic called LDA-value, to verify the correlation between different age groups and the LDA-value. Results: The results showed that 125 characteristics were able to discriminate differences between the young and elderly groups. The use of the LDA-value allowed obtaining a linear model of the changes that occur with aging in the performance of tasks with advancing age, the correlation obtained using the Pearson coefficient was 0.86. Conclusion: When we compared only the groups of young people and elderly people, the results indicated that there is a difference in the way tasks are performed between young people and elderly people. When the 8 groups were analyzed, the linear correlation obtained was strong, with the LDA-value being effective in obtaining a linear correlation of the eight groups, demonstrating that although the characteristics alone did not demonstrate gradual changes depending on age, the combination of these highlighted the changes.