Can a smartwatch detect coronavirus?

Changes in resting heart rate can alert smartwatch wearers to potential infections before symptoms
13 December 2021


A smartwatch


Researchers at Stanford University’s Department of Genetics are harnessing the treasure trove of data that is being recorded 24/7 by smartwatches. Although wearable tech, including rings, wristbands and even sensors that can be incorporated into clothing, can do more than just measure heart rate and daily steps, it would appear that this simple data alone can provide powerful insight into the real-time health status of its wearer.

Mike - We're following your normal, healthy baseline. Then if there's a jump up in signal, an alert goes off and it works about 80% of the time. We'll get an alert at or before a symptom and it turns out it's an average of three days from our real time alerting system. So three days prior to people getting ill, on average a red alert will go off. Now we do miss it in 20% of the cases and we think that's because it's hard to get a stable, healthy baseline for some people. That makes finding these triggers a lot harder. That's why we're trying to improve our algorithms as well.

Tricia - And all of that is at the moment done from resting heart rate, but you want to look at heart rate variability, activity level and sleep quality. What are you hoping to find from that additional data?

Mike - We think that will be key to telling the difference between say a respiratory viral infection and a mental health stress, and other sorts of triggers. Things like alcohol and intense exercise can all trigger these alerts. Most of them you can contextualize and simply ignore. Certainly intense exercise, we should be able to ultimately subtract out because we'll see that from the smart watch.

Tricia - Yeah. Because with intense exercise, you're probably gonna have some kind of evidence of elevated heart rate from the intense exercise itself.

Mike - Exactly.

Tricia - What about asymptomatic cases? Were you able to detect those and how far in advance did you detect them?

Mike - Remarkably, we can find asymptomatic covid cases as well. 14 out of 18 asymptomatic cases we could pick up and in some cases, the signal appeared two weeks prior to when they were diagnosed.

Tricia - Because I'm a religious tracker of my own health, I already have access to my heart rate data and I tested positive asymptomatically and my heart rate spiked. Then it went down again at roughly the same time. But it's interesting that you say the alert comes about two weeks prior to that in some asymptomatic cases, because I had another infection about two weeks prior to that.

Mike - In our very first study, what we noticed is that about half the time people had another stress event prior to their COVID. We don't know what that was due to, it could have been another illness, it could have been possibly a mental stress, it's not clear. But we've noticed it was quite frequent, more than you would expect at random. What that means is that we think there may be early stressors that perhaps make people more susceptible to viral infections as well.

Tricia - You're preaching to the choir a little bit because I love my data and I wear a smartwatch. But if you were to try and convince someone who didn't really track their health and didn't wear an apple watch or anything like that, what would you say to them to convince them to start monitoring their health in real time?

Mike - I'd say, 'do you drive your car around without a dashboard?' Why do you drive your body around without a health dashboard? It makes no sense. I think that's what we need for human health. We currently rely on internal sensors, which kind of work, but they're slow. We think that these physiological sensors from smart devices can actually find things much, much earlier and warn people ahead of time before symptoms and that's super powerful.

Tricia - Every year, these smart watches seem to add another sensor. What do you think is the future for wearable technology and how much do you think we're gonna be able to gain from those wearable devices?

Mike - I think we're only at the beginning. For example, there's something called 'Galvanic Stress Response' that measures the conductance on your skin. That actually can tell you when you're getting stressed, because the more you sweat, the more it'll pick up a signal. Likewise, if you have diabetes, your skin gets drier and that can get picked up as well. So it's making certain kinds of measurements that we don't currently do and so we can actually incorporate that into health monitoring in a way that's never been done.

Tricia - So you're saying that nowadays my watch tells me to move if I haven't moved in an hour, but in the future, it's gonna tell me to drink and eat as well.

Mike - I think so. And it'll tell you when you're getting stressed, so maybe you better sit down and meditate.

Tricia - And it'll guide me through meditation as well.

Mike - Probably. Play music automatically. Maybe that's where we're going.


Add a comment