Research on COVID detection smartwatches has not yet been successful

In the early days of the COVID-19 pandemic, smartwatch advocates and wearable technology companies believed that devices could help detect illness. They wanted to use things like heart rate and oxygen levels to flag people who might be ill. This strategy may still be a reasonable way to track down the illness, but two years later, the promise has not been fulfilled. The study has not yet been developed, according to a new review published last week. Lancet..

This review explores 12 research studies and 12 proposed research protocols published in 2020 and 2021 that attempted to find patterns in data collected by devices such as the Apple Watch, Fitbit, and Whoop. rice field. Most of these studies focused on people who were already positive for COVID-19 testing. Instead of tracking healthy people to predict who will get sick, researchers looked for patterns in wearable data a few days before they got sick. The authors of this new study state that none of the studies were rigorous clinical trials. There are no existing studies tested to see if wearable devices actually lead to early detection of COVID-19.

Studies have shown that most algorithms built to screen COVID-19 from wearable data focus primarily on symptomatic treatment. The four tried to detect the infection before the person began to show symptoms, but the success varied. We were able to detect 20-88% of infections. The model became less accurate as the number of days before trying to predict the disease increased. “The accumulated evidence suggests a trade-off between the accuracy of the model and the ability to identify SARS-CoV-2 infection before the onset of symptoms,” the review author writes. It also reduces the usefulness of the device as a COVID-19 detector — some of the promises of this type of strategy are to get sick fast enough to be tested and isolated before the disease spreads to others. Is to flag people who are.

There is ample evidence that physiological signals such as changes in body temperature, fluctuations in heart rate, and other metrics are associated with someone getting sick. However, this included not only COVID-19 but also other illnesses, and most studies in this review did not distinguish between COVID-19 and something like the flu. A study conducted at Fitbit found duplicates between data on influenza and data on COVID-19, said Conor Heneghan, head of research at Fitbit. The Verge last year. “My instinct is that it will be difficult to distinguish them reliably,” he said.

The authors of the new review write that using wearable devices as detectors for COVID-19 or other illnesses also poses a fairness issue. Due to the lack of racial diversity in the studies included in the analysis, it is not clear whether the model works equally well in non-white populations. This is even more of a concern, as research shows that wearables behave differently in dark skin tones and are often less accurate. Also, none of the models evaluated in the review considered the menstrual cycle, even if there were changes in other variables related to different stages of body temperature and cycle.

Although existing research has limitations, wearables can be a great way to track and monitor illness. Better research needs to be done to prove it and find the best way to use the device in these situations. Experts believe that any basic tool that can warn someone that they may be ill will still help.

Jennifer Ladin, an epidemiologist in the Digital Medicine Division of the Scripps Research Translational Institute, said: The Verge last year..

Leave a Comment