Wearable activity trackers combined with AI may aid in early identification of COVID-19

Wearable activity trackers that track changes in skin temperature and heart and respiratory rates, combined with artificial intelligence (AI), could be used to pick up COVID-19 infection days before symptoms start, preliminary research published in the open access journal suggests. magazine BMJ Open

The researchers base their findings on wearers of the AVA bracelet, a regulated and commercially available fertility tracker that monitors respiratory rate, heart rate, heart rate variability, wrist skin temperature and blood flow, as well as sleep quantity and quality.

Typical COVID-19 symptoms can take several days to appear after infection, during which time an infected person can inadvertently spread the virus.

Attention has begun to focus on the potential of activity trackers and smartwatches to detect all stages of COVID-19 infection in the body, from incubation to recovery, with the aim of facilitating early isolation and testing of people with the infection.

The researchers therefore wanted to see if physiological changes, followed by an activity tracker, could be used to develop a machine learning algorithm to detect COVID-19 infection before symptoms start.

Participants (1163 all under the age of 51) were drawn from the GAPP study between March 2020 and April 2021. GAPP, which started in 2010, aims to better understand the development of cardiovascular risk factors in the general population of Lichtenstein.

The AVA bracelet was chosen because the data was previously used to inform a machine learning algorithm to detect the most fertile days of women’s ovulation in real time, with 90% accuracy.

Participants wore the AVA bracelet at night. The device stores data every 10 seconds and requires a minimum of 4 hours of relatively uninterrupted sleep. The bracelets were synchronized with an additional smartphone app upon awakening.

Participants used the app to record all activities that could potentially alter central nervous system functioning, such as alcohol, prescription and recreational drugs, and to record possible COVID-19 symptoms.

They all did regular rapid antibody tests for SARS-CoV-2, the virus responsible for the COVID-19 infection. Those with indicative symptoms also took a PCR smear.

Everyone provided personal information about age, gender, smoking status, blood type, number of children, exposure to household contacts or colleagues who tested positive for COVID-19 and vaccination status.

About 127 people (11%) developed a COVID-19 infection during the study period. There were no differences in background factors between those who did and did not test positive. But a significantly higher proportion of those who did said they had been in contact with family members/regulars or colleagues who also had COVID-19.

Of the 127 who tested positive for COVID-19, 66 (52%) had worn their bracelet at least 29 days before the onset of symptoms and were found positive by a PCR smear, so they were included in the final analysis.

The monitoring data revealed significant changes in all five physiological indicators during the incubation, presymptomatic, symptomatic and recovery periods of COVID-19 compared to baseline measurements. The symptoms of COVID-19 lasted an average of 8.5 days.

The algorithm was ‘trained’ using 70% of the data from day 10 to day 2 before symptom onset within a 40-day period of continuous monitoring of the 66 people who tested positive for SARS-CoV-2. It was then tested on the remaining 30% of the data.

Approximately 73% of lab-confirmed positives were picked up in the training set and 68% in the test set up to 2 days before symptom onset.

The researchers acknowledge that their results may not be more broadly applicable. The findings were based on only a small sample of people, all of whom were relatively young — thus less likely to have severe COVID-19 symptoms — from a single national center, and who were not ethnically diverse.

In addition, the achieved accuracy (sensitivity) was less than 80%. But the algorithm is now being tested on a much larger group (20,000) people in the Netherlands, with results expected later this year, they say.

While a PCR smear remains the gold standard for confirming COVID-19 infection, “our findings suggest that a wearable informed machine learning algorithm could serve as a promising tool for presymptomatic or asymptomatic detection of COVID-19,” they write. .

And they conclude: “Wearable sensor technology is an easy-to-use, low-cost method to enable individuals to monitor their health and well-being during a pandemic. Our research shows how these devices, when combined with artificial intelligence, can push the boundaries of personalized medicine. and detecting diseases prior to: [symptom occurrence]potentially reducing the transmission of viruses in communities.”


Reference magazine:

Risch, M., et al. (2022) Investigating the use of a sensor bracelet for the presymptomatic detection of changes in physiological parameters related to COVID-19: an interim analysis of a prospective cohort study (COVI-GAPP). BMJ Open. doi.org/10.1136/bmjopen-2021-058274.

Leave a Comment

Your email address will not be published.