Purdue University is committed to in-person learning, and the school is leaning on the brainpower of its researchers to find cutting-edge ways to manage the pandemic on campus. A new study is underway—with mostly Boilermakers enrolled—to see if smartwatches could provide the earliest clues that the wearer could be developing COVID-19, even before the first hint of a symptom. The work is the first step toward a possible method to monitor students’ health en masse; they could be handed a smartwatch, much like a basic school supply, or perhaps use their own device to provide their biometric data as an early indicator that they might be getting sick.
“If you look at what Purdue is investing now with [COVID-19] testing, quarantining space—everything—a smartwatch is a tiny fraction of what the university has already invested,” says Purdue’s Leslie A. Geddes Associate Professor of Biomedical Engineering Dr. Craig Goergen. “Everybody has a smartphone now, and nobody thought about that 20 years ago being a possibility.”
Scientists have already shown that changes in a person’s heart rate and breathing rate—even before symptoms appear—could indicate the person is coming down with COVID-19. For example, if an individual’s resting heart rate is five to 10 beats faster than usual, it could indicate “that your body’s natural state is kind of revved up,” says Goergen.
Goergen is leading a study at Purdue to see if smartwatches could be used to reliably capture changes in heart rate and breathing rate that would indicate a viral infection, such as COVID-19. About 60 people—mostly on campus—have already participated in the study, and the team aims to enroll 100 total. Each person is given a smartwatch to measure heart rate and other biometric data and also wears a chest patch that’s widely accepted and approved by the FDA to collect an electrocardiogram signal.
“The chest patch has a great level of accuracy in its ability to measure heart rate and respiration rate,” says Goergen.
The scientists will compare the “gold standard” data collected by the chest patch to the data collected by the smartwatch to see if the smartwatch is able to accurately capture heart rate and breathing rate. The Purdue team is partnering with Chicago-based physIQ, founded by Purdue graduate Gary Conkright, which has expertise in data analytics and algorithms for biometric data.
physIQ will use its FDA-cleared artificial intelligence-based algorithms to analyze the heart rate, respiration rate and heart rate variability collected by the chest patches the participants wear in the study. The Purdue Research Foundation’s Foundry Investment Fund recently invested $500,000 in physIQ, which employs a significant roster of Purdue alumni.
Goergen says Purdue faculty “will take the data [collected by the smartwatches] and run with it” to develop algorithms that would help smartwatches or activity trackers better measure heart rate, respiration rate and heart rate variability. Because each person has a unique resting heart rate, the smartwatch will establish the person’s unique baseline, so the data could recognize a change.
Goergen expects the study results will be consistent with similar research and show significant change in subjects’ data using the biometric analysis approach. The study is funded by a faculty innovations grant from Protect Purdue, the school’s initiative to keep the campus and surrounding community safe from COVID-19.
“We’re not at the point now where we can roll this out to all faculty, staff and students next week, but our hope is this is a step forward in the right direction,” says Goergen. “Whether it’s this pandemic or the next, we’ll be more prepared and can help with the process to limit spread, limit infection numbers and help people identify when they maybe have developed a viral infection or be in a state where they’re infectious.”
Goergen says in a matter of years, students could potentially be given a smartwatch or use another activity tracker to monitor these biometric indicators of infection. He believes, in the future, there will be wearable devices that are inexpensive and reliable enough to consider for mass distribution.
“There might be a Purdue app that says, if you opt in and you want to be part of this program, we can help with AI-based monitoring, where your biometric data—which won’t be shared and spread to people you don’t want it to—can be tracked and monitored. So if you do get sick, we can help identify that early; you may need to undergo early testing or visit the clinic to help prevent spread,” says Goergen. “We’re trying to figure out how to help and what to do to make a difference. We still have a long way to go, but even if this is a small step in that direction, we’re really excited about it.”
Goergen explains what his research team expects to find and how it opens the door to other applications using smartwatches.
Goergen says some parts of the study have already been “a smashing success,” and the next step is focusing on people who have had COVID-19 or are at-risk.