Wearable technology has the power to transform everyday fitness trackers into powerful tools that do more than just count steps. Scientists are now harnessing these gadgets to detect subtle signs of diseases, potentially spotting health issues before symptoms even emerge—keeping a digital finger on the pulse of our well-being.
Varsha Gupta, a Senior Scientist at A*STAR Institute for Human Development and Potential (A*STAR IHDP), previously known as the Singapore Institute for Clinical Sciences (SICS); and Bioinformatics Institute (BII), noted that healthcare monitoring became crucial during the COVID-19 pandemic as researchers found a connection between physical activity and prolonged COVID-19 symptoms in healthcare workers.
“COVID-19 symptoms were taking observably differing lengths of time to resolve. At the same time, movement and social distancing restrictions may have had an impact on daily life physical activity patterns,” Gupta said. This observation prompted the team to investigate whether recorded physical activity levels were linked to COVID-19 symptoms.
Together with Senior Principal Scientist Dennis Wang (from A*STAR IHDP and BII), Gupta collaborated with researchers from the University of Sheffield, UK and Stanford University, US, on a year-long study of staff at the UK’s National Health Service (NHS). The team analysed data from NHS healthcare workers who recorded their symptoms via an app and tracked their physical activities with Apple watches starting from April 2020.
Gupta explained that collecting data in real-life settings introduces variability because participants have different daily routines and sometimes forget to wear their watches. As a result, the team had to implement rigorous data cleaning processes to handle self-reported data, removing outliers and duplicate entries.
Using a data-driven approach, the researchers applied unsupervised machine learning techniques to identify the evolution of COVID-19 symptoms and physical activity patterns among the participants. They discovered a correlation between the distances participants walked or ran and the duration of their COVID-19 symptoms: those with prolonged symptoms generally showed reduced physical activity from the infection's onset.
These findings highlight the persistence of symptoms in mild COVID-19 cases, as well as what Gupta sees as a promising future for wearable technology in shaping healthcare. “The temporal patterns of markers from smartwatches can provide a deeper understanding of the dynamic nature of lifestyle-health associations, enabling us to identify intervention points and novel insights,” said Gupta.
These findings also suggest that future studies should consider differences in device brands, user lifestyles and varying diseases to fully harness the technology's potential, Gupta added. The team is continuing their research, working with international partners to explore further how health markers relate to physical activity across different settings.
The A*STAR-affiliated researchers contributing to this research are from the A*STAR Institute for Human Development and Potential (A*STAR IHDP) and Bioinformatics Institute (BII).
