Michigan Researchers Use a Smartwatch to Detect, Monitor COVID-19

Researchers at the University of Michigan have developed algorithms that allow them to track the progress of the virus through a smartwatch, and hope to expand that platform to monitor other health concerns.

Researchers at the University of Michigan have created a protocol for tracking COVID-19 symptoms through a smartwatch, and say the process could eventually be used to detect other health issues, such as the flu.

In a study recently published in Cell Reports Medicine, the research team traced six factors derived from heart rate data collected by a smartwatch that determine when a user is infected by the virus and how sick they become. They found that those living with the virus experienced an increase in their heart rate per step once the symptoms were detected, and those dealing with a cough experienced a much higher heart rate per step than those who didn’t have a cough.

“We found that COVID dampened biological timekeeping signals, changed how your heart rate responds to activity, altered basal heart rate and caused stress signals,” Daniel Forger, a professor of mathematics and research professor of computational medicine and bioinformatics at the University of Michigan and part of the research team, said in a press release issued by the university. “What we realized was knowledge of physiology, how the body works and mathematics can help us get more information from these wearables.”

The research adds to the growing body of evidence that mHealth wearables can be used to detect and monitor COVID-19 in patients at home, enabling healthcare organizations to treat them through remote patient monitoring programs rather than putting them in a hospital.

It also expands the opportunities for RPM programs to track and treat other health concerns at home instead of the hospital, clinic or doctor’s office. In time, wearables and sensor-embedded clothing could be used to detect and monitor a wide range of health concerns, from viruses like the flu to chronic conditions like diabetes, asthma, cardiac failure and cancer.

In Michigan, researchers focused on data from patients in the Intern Health Study, a multi-site study that followed physicians in several locations across their first year of residency, as well as the Roadmap College Student Data Set, which tracked student health during the 2020-21 school year through Fitbit wearables, self-reported COVID-19 diagnoses and symptom information and publicly available data. In all, they tracked the health of 43 medical interns and 72 students.

Using an algorithm developed to estimate daily circadian phase from heart rate and step data taken from a wearable, they found that:

• Heart rate increase per step, a measure of cardiopulmonary dysfunction, increased after symptom onset.
• Heart rate per step was significantly higher in participants who reported a cough.
• Circadian phase uncertainty, the body’s inability to time daily events, increased around COVID symptom onset. Because this measure relates to the strength and consistency of the circadian component of the heart rate rhythm, this uncertainty may correspond to early signs of infection.
• Daily basal heart rate tended to increase on or before symptom onset. The researchers hypothesize this was because of fever or heightened anxiety.
• Heart rate tended to be more correlated around symptom onset, which could indicate the effects of the stress-related hormone adenosine.

The research team said they were able to create new algorithms that can be used to study how an illness impacts heart rate physiology – and which could be used to expand the use of wearables in healthcare.

“There’s been some previous work on understanding disease through wearable heart rate data, but I think we really take a different approach by focusing on decomposing the heart rate signal into multiple different components to take a multidimensional view of heart rate,” Caleb Mayer, a doctoral student in mathematics, said in the press release. “All of these components are based on different physiological systems. This really gives us additional information about disease progression and understanding how disease impacts these different physiological systems over time.”