Sensores inerciais móveis no rastreio e predição de risco de quedas em idosos

• Ano de defesa: 2019
• Autor(a) principal: Bet, Patrícia
• Orientador(a): Ponti, Moacir Antonelli
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de São Carlos; Câmpus São Carlos
• Programa de Pós-Graduação: Programa de Pós-Graduação em Gerontologia - PPGGero
• Palavras-chave em Português: Aceleração; Análise de marcha; Risco de quedas; Sensores vestíveis; Idosos; Elderly
• Palavras-chave em Inglês: Acceleration; Gait analysis; Fall risk; Wearable sensors
• Área do conhecimento CNPq: CIENCIAS DA SAUDE::SAUDE COLETIVA::MEDICINA PREVENTIVA
• Link de acesso: https://repositorio.ufscar.br/handle/20.500.14289/10897

Background: The investment in acceleration sensor-based equipment as a fall prevention strategy can represent a viable option for the monitoring of the fall risk. However, more studies are needed on the analysis of such data, in particular for the prediction of future falls to be adopted for health care and prevention. Thus, the objective of this study is to investigate patterns for fall risk screening in the elderly with no recent history of fall, through three types of Timed Up and Go (TUG) tests using a single acceleration sensor. Methods: A prospective study was conducted with 74 healthy elderly non-fallers using waist acceleration sensors while performing three variations of the TUG. After evaluation at baseline, the occurrence of falls (outcome) was monitored quarterly during one year. We investigate frequency features extracted from the accelerometry signal and their ability to predict falls. Mann-Whitney U test was used to compare the groups. Besides, analyze of accuracy, sensitivity and specificity were performed. Results and Discussion: The best individual feature result shows an accuracy of 0.75, sensitivity of 0.71 and specificity of 0.76. A fusion of the three best features increases the sensitivity to 0.86. Statistical difference was found in the whole accelerometry data, without any type of processing, for three, six, nine and twelve months of follow-up. The cut-off points of the TUG seconds did not demonstrate adequate sensitivity for community-dwelling elderly. Conclusions: The results confirms previous evidence that accelerometer features can better fall risk, and support potential applications that try to infer falls risk in less restricted scenarios, even in a sample stratified by age and gender composed of active community-dwelling elderly.