Health-Related Physical Fitness Level From Smartwatches

Health-related physical fitness (HRPF) has demonstrated high clinical relevance, and its level is associated with the ability to perform activities of daily living with vigor and a lower risk of chronic disease. Consequently, exercise prescription guidelines recommend improving HRPF as a focus for prevention and rehabilitation programs. Measuring and tracking HRPF often requires specialized equipment and personnel, which are expensive and less applicable to the general population. Wearables may mitigate this issue by providing useful estimates of the HRPF.

Health-related physical fitness (HRPF) has high clinical relevance \[1\]. It is associated with the ability to perform activities of daily living with vigor and a lower risk of chronic disease \[2\]. Consequently, exercise prescription guidelines recommend improving HRPF as a focus for prevention and rehabilitation programs \[3\]. The American College of Sports Medicine (ACSM) \[3\] grouped the HRFP into five domains: cardiorespiratory endurance, body composition, muscular strength, muscular endurance, and flexibility. However, measuring and tracking the fitness levels for all HRPF domains requires specialized laboratory equipment and personnel, which are expensive and less applicable to the general population. Wearable technology mitigates this issue and has proven to be a reliable alternative capable of providing useful estimates of the HRPF \[4\] \[5\] \[6, 7\]. Previous work has predicted ACSM HRPF domains from anthropometric and laboratory bioelectrical impedance analysis data (BIA) \[8\] \[9\]. Nevertheless, their data are based on the National Fitness Award (NFA), a nationwide test used to assess the physical fitness of the general South Korean population that is collected using specialized laboratory equipment under the supervision of health professionals. Current advances in wearables may allow us to estimate the fitness level for all HRPF domains using only smartwatch data, enabling economic, non-intrusive predictions and being available during the user's daily routine. The complete characterization of health-related fitness as a multidimensional depiction of the user's fitness status can be used to track health status continuously and to design specialized training prescriptions. The main goal of this study is to estimate the fitness level for all HRFP domains from data obtained from smartwatches during unsupervised activities of daily living. We hypothesized that data from smartwatches could be used to estimate the fitness levels from all HRPF domains.