Toward characterizing cardiovascular fitness using machine learning based on unobtrusive data

Autor(es)

Frade, Maria Cecilia Moraes ; Beltrame, Thomas ; Gois, Mariana de Oliveira ; Pinto, Allan ; Tonello, Silvia Cristina Garcia de Moura ; Torres, Ricardo da Silva ; Catai, Aparecida Maria ; Jaafar, Zulkarnain

• Tipo de documento: Artigo Científico
• Fonte: PLoS One; v. 18, n. 3, p. 18-pg., 2023-03-02.

Resumo

Cardiopulmonary exercise testing (CPET) is a non-invasive approach to measure the maximum oxygen uptake (V O2-max), which is an index to assess cardiovascular fitness (CF). However, CPET is not available to all populations and cannot be obtained continuously. Thus, wearable sensors are associated with machine learning (ML) algorithms to investigate CF. Therefore, this study aimed to predict CF by using ML algorithms using data obtained by wearable technologies. For this purpose, 43 volunteers with different levels of aerobic power, who wore a wearable device to collect unobtrusive data for 7 days, were evaluated by CPET. Eleven inputs (sex, age, weight, height, and body mass index, breathing rate, minute ventilation, total hip acceleration, walking cadence, heart rate, and tidal volume) were used to predict the V O(2-max)by support vector regression (SVR). Afterward, the SHapley Additive exPlanations (SHAP) method was used to explain their results. SVR was able to predict the CF, and the SHAP method showed that the inputs related to hemodynamic and anthropometric domains were the most important ones to predict the CF. Therefore, we conclude that the cardiovascular fitness can be predicted by wearable technologies associated with machine learning during unsupervised activities of daily living. (AU)

• Processo FAPESP: 18/19016-8 - Associações entre aptidão e potência aeróbia
• Beneficiário: Thomas Beltrame
• Modalidade de apoio: Bolsas no Exterior - Estágio de Pesquisa - Pós-Doutorado
• Processo FAPESP: 16/22215-7 - Impacto do treinamento muscular inspiratório e do envelhecimento no mapeamento metabólico, modulação autonômica e respostas cardiovasculares, respiratórias e metabólicas e predição da saúde cardiorrespiratória por meio de wearables
• Beneficiário: Aparecida Maria Catai
• Modalidade de apoio: Auxílio à Pesquisa - Temático
• Processo FAPESP: 17/09639-5 - Explorando técnicas de aprendizado de máquina para a modelagem do sistema aeróbio e suas aplicações em programas de reabilitação cardiorrespiratória
• Beneficiário: Thomas Beltrame
• Modalidade de apoio: Bolsas no Brasil - Pós-Doutorado
• Processo FAPESP: 18/22818-9 - Wearables para a monitorização contínua da saúde cardiorrespiratória
• Beneficiário: Maria Cecília Moraes Frade
• Modalidade de apoio: Bolsas no Brasil - Doutorado
• Processo FAPESP: 19/16253-1 - Desvendando o segredo do futebol Brasileiro e Holandês, capturando elementos de estilo de jogo e estratégias de sucesso
• Beneficiário: Allan da Silva Pinto
• Modalidade de apoio: Bolsas no Brasil - Pós-Doutorado