In PeerJ
In recent years, many people have been working from home due to the exceptional circumstances concerning the coronavirus disease 2019 (COVID-19) pandemic. It has also negatively influenced general health and quality of life. Therefore, physical activity has been gaining much attention in preventing the spread of Severe Acute Respiratory Syndrome Coronavirus. For planning an effective physical activity for different clients, physical activity intensity and load degree needs to be appropriately adjusted depending on the individual's physical/health conditions. Heart rate (HR) is one of the most critical health indicators for monitoring exercise intensity and load degree because it is closely related to the heart rate. Heart rate prediction estimates the heart rate at the next moment based on now and other influencing factors. Therefore, an accurate short-term HR prediction technique can deliver efficient early warning for human health and decrease the happening of harmful events. The work described in this article aims to introduce a novel hybrid approach to model and predict the heart rate dynamics for different exercises. The results indicate that the combination of singular spectrum analysis (SSA) and the Clayton Copula model can accurately predict HR for the short term.
Namazi Asieh
2022
Artificial intelligence, Heart rate, Machine learning, Prediction, Wearable sensors
