Wearable health trackers can help spot COVID-19 days before symptoms appear, according to a recent study.
Published in the journal BMJ Open(Opens in a new window), the results suggest that combining physiological data with machine learning can lead to diagnoses of SARS-CoV-2 infections ahead (or in the absence) of tell-tale signs.
"Wearable sensor technology can enable COVID-19 detection during the presymptomatic period," a team of international researchers concluded.
The group—hailing from the Dr. Risch Medical Laboratory in Liechtenstein, University of Basel in Switzerland, McMaster University in Canada, and Imperial College London—used the Ava wearable, which records respiratory rate, heart rate, HR variability, wrist skin temperature, and blood flow overnight.
In total, 1,163 people under the age of 51 wore the fertility tracker (which retails for $279(Opens in a new window)) from April 2020 through March 2021. The trackers were synchronized with a smartphone app, allowing people to record behaviors that may interfere (alcohol, medication, recreational drugs), as well as possible COVID symptoms.
A total of 1.5 million hours of physiological data were recorded and a COVID-19 infection was confirmed in 127 people. 66 (52%) of those infections were for individuals who wore their device for at least 29 consecutive days. The study found "significant" changes in the body before, during, and after infection—including higher respiration and heart rates—compared to those who tested negative.
The team's combination of health tracker and algorithm, designed to facilitate early isolation and testing of potentially affected people and limit the spread of COVID, correctly identified 68% of positive cases two days before the onset
Read more on pcmag.com