Researchers have found out in preliminary studies that the data collected by wearables can be used to discover corona virus symptoms a lot earlier than you realize you are sick. This would potentially make fitness trackers, sickness trackers.

Initial findings from two academic studies constitute a small step in the war against corona virus and a large jump for wearable gadgets. In case Fitbits, Apple Watches and Oura smart rings turnout to be useful in warning about possible corona virus infection early, they could very much facilitate reopening of workplaces and communities, and develop into health necessities from simply consumer tech.

Many studies since March, have been constantly trying to discover whether the data collected by the wearables about our bodies provide any evidence about a potential case of corona virus. Fitbit has also announced its own study.

Product laydown photography for Fitbit Charge 4.

The greatest promise, however, maybe shown by a not so much popular wearable called Oura ring, which is made by a seven year old company based in Finland and the United States. Costing $300, the ring looks much like jewellery and collects data about breathing, heart rate, and temperature—which is vital for corona virus. Presently, this ring is being used in two studies: at the University of California in San Francisco and at the West Virginia University, involving thousands of health care workers.

Although, such studies are yet to publish conclusive results, they might be producing the first evidence that this idea indeed works. Researches at West Virginia University’s Rockefeller Neuroscience Institute described that the data collected by Oura ring together with an app to measure cognition can predict up to three days earlier, when people will show signs of fever, coughing or shortness of breath.

The institute’s director, Professor Ali Rezai, has stated that this technology is extremely important to discover infection early on, when patients are highly contagious but are unaware. He considers the combination of wearables and app a type of “digital PPE”. ‘It can say, “This individual needs to stay home and not come in and infect others,’” he said.

Furthermore: researchers at Stanford said that they have been successfully able to predict corona virus at the time of diagnosis or earlier, in 9 out 14 confirmed patients they have studied. In one case, they discovered a jump in the patient’s heart rate 9 days before the patient reported any symptoms. However, in other cases, the data provided evidence of infection in the patients, when the patients themselves noticed symptoms. “The bottom line is it is working, but it’s not perfect,” said Stanford professor Michael Snyder.

Given the hysteria that often engulfs consumer technology products, there is plenty of reasons to be skeptical about technology charting an uncertain course towards an illness that in many respects is still a mystery. Researchers also need to crunch more numbers to determine the difference between a coronavirus patient and another case.

According to conservative scientists, we are still weeks or months away from developing warning systems against the disease, that can be clinically tested.

Ben Smarr, a professor at the University of California at San Diego who is helping lead the data-crunching on the UCSF study said, “I haven’t seen that subtlety embraced by most tech companies,”. Furthermore, he stated, “I’m wary because I don’t want this to be used to sell people a false solution or false hope.”

 

Counting steps to predicting infections

The main question that arise in the quest to successfully detect corona virus using gadgets is: accuracy.

Fitness wearables initially aimed at tracking the number of steps taken by the user. However, lately, tech companies have been keen in health care, hence adding more sensors to these gadgets. Fitbits are now collecting heart data, and in the previous year, google decided to acquire it, as a means to get close to the bodies of its consumers. In a first, Apple got Food and Drug Administration clearance for a watch that could detect atrial fibrillation.

Previous studies have been successful to show the link between wearables and undetected issues like arrythmia , high-blood issues, early stage cancer and their use to improve real-life tracking of seasonal outbreaks of flu.

Corona virus can be a real game changer according to researchers. A professor at Duke University, Ryan Shaw stated “Because everybody is going through this, it is an opportunity for us to collect data from essentially the entire population, which is very unique”.

The question, however, arises is that how do we use these devices that are not designed for medical purposes and obtain health information from data collected by them? Wearables researchers say that they treat this data as a baseline and not as an individual measurement; to identify variation based on data that is considered as normal behaviour for the body.

The researchers then input weeks of data into software that can identify patterns. The algorithms in these software can identify details that humans usually cannot, such as slightly increased heart rate. Subtle temperature changes, heart rate variation and sleep patterns allow the software to extrapolate what will happen in the coming days.

The studies at Duke, Stanford and Scripps are welcoming to data coming from any wearable that the users own. However, there are questions about the accuracy of data collected by some these products. “We don’t believe that any of the devices that we’re using in our study are bad enough that we wouldn’t be able to capture the signals that we expect to capture,” Duke professor Jessilyn Dunn.  `

Researchers say that the Oura ring has several advantages over other smart watches, even though it is not as popular as other wearables with about 150,000 users.

The Oura ring is very small in size, this means that users are more likely to wear it while sleeping, compared to other wearables, and the best time to measure resting heart rate is while a person is asleep.

According to a nurse manager in Morgantown, W.Va., Sara Belch, she wears her Oura ring 24 hours a day, and it only needs to be charged after 4 or 5 days. She said “It’s smooth, and I don’t feel any different wearing it,”.

Additionally, the Oura ring can record temperature from the finger, which is an important feature lacking in other wearables. TemPredict is a theory that is being tested by UCSF study, and it states that people with latent coronavirus show variations in body temperature that can be detected through constant monitoring.

Oura donated rings to UCSF study and also gave access to raw data from participants to the researchers of both UCSF study and RNI, to facilitate the research.

Another key factor is the data that comes not from the wearables, rather comes directly from the participants. Participants in all of the studies have been asked to constantly check in through apps and websites to report symptoms such as coughing or the result of coronavirus test they underwent.

Perhaps, the most demanding study of all is the RNI study, in which participants are asked to check into its special app two times a day, which also includes participation in games that check the participant’s attention and other brain functionalities. Additionally, participants have to measure and report their temperature which is often verified by a professional.

RNI claim that their software is 90% accurate in predicting the beginning of coronavirus symptoms. This, however, is based on the group that RNI has studied— about 600 health care workers and first responders.

Rezai says that in order to detect coronavirus instead of just symptoms, more participants are required to train the algorithms so that they can detect the different and often unexpected ways that different bodies react to the coronavirus.

 

Converting the research into an effective early warning system

The notion that wearables provide useful data is widely agreed upon. However, turning this data into an effective early warning system against coronavirus, has many challenges.

One of those challenges is privacy. In order to participate in one of these researches, a person must agree to constant monitoring of the body. The Oura ring, for example records heart rate and temperature every second.

The next move will be for the researchers to perform trials where they send real-time warnings to the participants of the study based on their own data. No one has done this so far since they are restricted by the laws of academic study. Snyder is awaiting Stanford permission to begin delivering input to a limited group of participants in the next few weeks. The UCSF study could achieve this milestone by this fall according to Smarr.

FDA regulates medical diagnostics, and hence its permission may be required if such systems are to be made available to the millions of the general population.

It is a complicated situation. Would the app be making recommendations regarding lifestyle based on the data? A bad diagnosis by the app can lead to serious consequences.

Wearable techs are currently negotiating with lots of disclaimers on that space. Oura reaches beyond others supplying people with guidance about how their bodies are shifting forward from day to day. It turns heart rate, sleep, activity and other data into a score of daily “readiness,” which suggests how much activity or rest you might need to optimize the day ahead.

For one Oura user this readiness score alone served as an early warning against coronavirus.

Petri Hollmén, a Finnish entrepreneur, who had been travelling in the European hotspots of coronavirus, was feeling well. One morning, however, his Oura app displayed a readiness score of 54, very low from his usual 80 to 90. The app had also noticed an increase in his body temperature of about 1 degree Celsius or 1.8-degree Fahrenheit.

This prompted him to take a coronavirus test—which turned out to be positive—even though he exhibited no symptoms.

This serves as cue for another extremely important factor about early warning tech: the information provided by the tech must be put to some use by us.

“It has to be paired with action like staying home and access to diagnostic tests to reduce false positives,” said Jennifer Radin, who is leading the Scripps study. “There are, of course, many other reasons why your resting heart rate can go up, like if you’re stressed or you change your exercise program.”