Stanford Medicine Scientists Hope to Use Data From Wearable Devices to Predict Illness, Including COVID-19
Stanford Medicine researchers and their collaborators, Fitbit and Scripps Research, are launching a new effort that aims to detect early signs of viral infection through data from smartwatches and other wearable devices.
By using wearable devices to measure things such as heart rate and skin temperature, which are known to elevate when the body is fighting off an infection, the team seeks to train a series of algorithms that indicates when your immune system is acting up.
If the algorithms succeed, the team hopes they could help curb the spread of viral infections, such as COVID-19.
“Smartwatches and other wearables make many, many measurements per day — at least 250,000, which is what makes them such powerful monitoring devices,” said Michael Snyder, PhD, professor and chair of genetics at the Stanford School of Medicine. “My lab wants to harness that data and see if we can identify who’s becoming ill as early as possible — potentially before they even know they’re sick.”
Snyder, who holds the Stanford W. Ascherman, MD, FACS, Professorship in Genetics, and his team are recruiting participants for the study through his lab’s Personal Health Dashboard. Fitbit, a company that makes wearable devices, will assist in that effort by raising awareness of the study with its users and offering them the option to participate. In addition, Fitbit plans to donate 1,000 smartwatches to Snyder’s research. As part of this collaboration, scientists at Scripps Research will also work with Fitbit to try to track how infection spreads in a community.
Once the algorithms are developed and verified, Snyder said, they could help people keep tabs on their health. Devices with an algorithm could alert users when their heart rate, skin temperature or some other part of their physiology signals that their body is fighting an infection. When people come down with a cold or flu, there’s usually a period just before symptoms set in when they wonder if they’re actually getting sick. Even during that time, without heavy symptoms, a sick individual often can still spread the virus. “You might wonder, ‘Are these sniffles allergies, or am I getting sick?’ These algorithms could help people determine if they should stay home in case their body is fighting off an infection,” Snyder said.
Watching for signs
Snyder’s research will be based on an algorithm that he and former postdoctoral scholar Xiao Li, PhD, now an assistant professor in the Center for RNA Science and Therapeutics at Case Western Reserve University, created in 2017. The algorithm showed that it was possible to detect infection using data — specifically, data from a change in heart rate — from a smartwatch. Snyder’s study showed that specific patterns of heart rate variation can indicate illness, sometimes even while the individual is asymptomatic. Li is also a collaborating principal investigator in the current study.
“Smartwatches and other wearables make many, many measurements per day — at least 250,000, which is what makes them such powerful monitoring devices,” Michael Snyder said.
For this study, Snyder is collecting data from five different brands of wearable device, including a smart ring and a variety of smartwatches. Each participant will also fill out surveys that keep track of their health status. Snyder and his team will create five new algorithms — one for each of the different wearables — to potentially detect when someone is getting sick. How quickly they can develop and verify the algorithms will depend on the number of participants who sign up for the study, Snyder said.
Although he’s hopeful that these algorithms will be able to successfully flag a specific change in heart rate linked to viral infection, Snyder also foresees some kinks to work out. “It’s possible that the algorithms could detect an elevated heart rate, but the user could be watching a scary movie or participating in some other activity that naturally elevates heart rate,” he said. “An alert isn’t a direct diagnosis, and it will be important for folks to be able to contextualize their situation and use some common sense.” Snyder also adds that even as his team works to develop algorithms that can flag illness, the next step is to investigate whether those signals can be sorted to be able to differentiate between viruses.
The study is an example of Stanford Medicine’s focus on precision health, the goal of which is to anticipate and prevent disease in the healthy and precisely diagnose and treat disease in the ill.
“I feel confident based on our former study that we’ll be able to detect some signal of infection based off of the wearables’ data,” Snyder said. “And I’m hopeful that as our study picks up, we may even have the granularity to anticipate the severity of viral infection based on smart device data. This tool may end up being a plus for both diagnosis and for prognosis.”
Written by Hanae Armitage
October 1, 2020 Update. This article is a preprint and has not been peer-reviewed.
Wearable devices digitally measuring vital signs have been used for monitoring health and illness onset and have high potential for real-time monitoring and disease detection. As such they are potentially useful during public health crises, such as the current COVID-19 global pandemic. Using smartwatch data from 31 infected individuals identified from a cohort of over 5000 participants, we investigated the use of wearables for early, presymptomatic detection of COVID-19. From physiological and activity data, we first demonstrate that COVID-19 infections are associated with alterations in heart rate, steps and sleep in 80% of COVID-19 infection cases. Failure to detect these changes in the remaining patients often occurred in those with chronic respiratory/lung disease. Importantly the physiological alterations were detected prior to, or at, symptom onset in over 85% of the positive cases (21/24), in some cases nine or more days before symptoms. Through daily surveys we can track physiological changes with symptom onset and severity. Finally, we develop a method to detect onset of COVID-19 infection in real-time which detects 67% of infection cases at or before symptom onset. Our study provides a roadmap to a rapid and universal diagnostic method for the large-scale detection of respiratory viral infections in advance of symptoms, highlighting a useful approach for managing epidemics using digital tracking and health monitoring.