News | Wearables | March 14, 2017

Clinical Study Uses Consumer Wearables to Anticipate Illness

NIH researchers are able to track health changes in one patient for two years with data from seven different wearable devices

wearable biosensors, anticipate illness, track health changes, PLOS Biology study

Wearable devices used in this study. The colors for different human figures indicate the specific studies in which each person participated. Red figures represent participation in all five studies; grey figures represent participation in the activity and insulin studies; blue, the activity, insulin sensitivity, and inflammation studies; orange and yellow, activity and air flights; green and pink, inflammation; and purple, air flights. Image courtesy of Li, et al, Stanford University.

March 14, 2017 — Researchers supported by the National Institutes of Health have revealed the ability of wearable biosensors, similar to the Apple Watch or Fitbit, to detect physiological changes that may indicate illness, even before symptoms appear. The findings, published Jan. 12, 2017, in PLoS Biology, may open the door to new ways to manage and monitor health, especially for those with limited access to doctors or clinics.

Changes in heart rate, blood pressure and body temperature can reveal health issues, such as cardiovascular disease or infection. While these are evaluated at yearly checkups, without more frequent monitoring, diseases can go unnoticed and progress between doctor visits. Additionally, these parameters vary greatly over the course of the day and between individuals, so a one-time reading may not be representative or give enough information to make a personalized assessment.

Wearable biosensors, like fitness trackers and activity monitors, can now check these parameters regularly. Although many people have started using them for personal use, larger, more comprehensive studies are needed to investigate their potential in healthcare.

“You can’t really make any scientific conclusions from personal data,” said Grace Peng, Ph.D., director of the National Institute of Biomedical Imaging and Bioengineering (NIBIB) program in Computational Modeling, Simulation, and Analysis. “Here, the authors have taken an enormous of amount of data and made sense out of the entirety in a very systematic way using data science tools and analytical methods.”

The team of researchers, from Stanford University and Veterans Affairs Palo Alto, first followed one man’s results for the two years that he wore seven different devices. Each device measured a variety of parameters, including his level of activity, radiation exposure and levels of oxygen-bound hemoglobin in his blood, and recorded more than 250,000 measurements each day. Based on its accuracy, ability to measure three key variables — heart rate, skin temperature and movement — and easily accessible data, they selected a Basis brand device to monitor an additional 43 subjects for an average of five months each.

In the first individual, they assessed how physiological measurements fluctuate over the course of the day, revealing the importance of frequent measurements, as a once-daily reading might not provide the whole picture. Comparing the other participants, they noted a wide range of heart rates, indicating the need for personalized analysis. The results also showed that wearable biosensors can provide both frequent measurements and personalized analysis.

Next, they evaluated instances when the readings appeared abnormal. The participant who was monitored for two years had four periods in which his heart rate and skin temperature were unusually high. During three of those periods, he reported clinical symptoms like congestion; following one of the outlying periods, he was diagnosed with Lyme disease. “The wearable was actually able to predict Lyme disease even before the patient had any symptoms,” said Peng.

Three other participants became sick at some point during the monitoring period. In each instance, the participant’s heart rate spiked, compared to his or her individual average. Using their data, the team created an algorithm, called Change-of-Heart, to identify possible illness based on when heart rate deviated from its normal range.

“It was incredibly exciting that we could actually detect illness using just a consumer wearable device,” said Jessilyn Dunn, Ph.D., a postdoctoral research fellow at Stanford University and one of the lead authors on the paper. To prevent false alarms, Dunn said they will continue to refine the algorithm to identify only true outliers and take into account one’s level of individual variability. She does not envision Change-of-Heart as a diagnostic tool just yet, but rather as a red flag to tell people they may need to examine their stress levels, get more rest, or see a doctor.

Although not a means of diagnosis yet, the study did raise the possibility of someday using wearable devices to identify risk for type 2 diabetes. Because many of the participants were at risk of becoming diabetic, the team was able to demonstrate the physiological signals, like high daytime heart rate, that correlated with insulin resistance. The study also investigated how parameters change during airplane flights and the ability of the monitors to detect radiation exposure.

The 24/7 monitors might also provide a more representative assessment of general health. “We know that the day that you go into the doctor once a year, it might not be your typical day, so we can get a picture of what that really is,” said Dunn.

Additionally, the trackers can monitor patients in rural or low-income areas that may not have easy access to a doctor. “For those individuals who can’t easily come into a clinic, they’re still under a sort of health surveillance mechanism,” Dunn said. “There’s a really exciting potential for this type of technology to really revolutionize the healthcare model.”

The research was funded in part by NIH grants, TR001085, EB020405, and DK102556.

For more information: www.journals.plos.org/plosbiology

Related Content

World Heart Federation Launches Global Roadmap on Cardiovascular Disease Prevention Among Diabetics
News | Cardiac Diagnostics | September 04, 2019
At the European Society of Cardiology (ESC) Congress 2019 together with the World Congress of Cardiology, the World...
Insomnia Tied to Higher Risk of Heart Disease and Stroke

Image courtesy of the American Heart Association

News | Cardiac Diagnostics | August 19, 2019
People suffering from insomnia may have an increased risk of coronary artery disease, heart failure and stroke,...
Evolutionary Gene Loss May Help Explain Human Predisposition to Heart Attacks
News | Cardiac Diagnostics | July 29, 2019
The loss of a single gene two to three million years ago in our ancestors may have resulted in a heightened risk of...
U.S. Soldiers Have Worse Heart Health Than Civilians
News | Cardiac Diagnostics | June 06, 2019
Active duty Army personnel have worse cardiovascular health compared to people of similar ages in the civilian...
Late Dinner and No Breakfast Worsens Outcomes After Heart Attack
News | Cardiac Diagnostics | May 23, 2019
People who skip breakfast and eat dinner near bedtime have worse outcomes after a heart attack, according to research...
HRS Releases New Expert Consensus Statement on Arrhythmogenic Cardiomyopathy
News | Cardiac Diagnostics | May 14, 2019
The Heart Rhythm Society (HRS) released a first-of-its-kind consensus statement with guidance on the evaluation and...
New Best Practices Help Manage Heart Attack Patients Without Significant Signs
News | Cardiac Diagnostics | April 15, 2019
For the first time in the United States, doctors with the American Heart Association (AHA) have outlined best practices...
The most recent U.S. Food and Drug Administration (FDA) clearance was Siemens Healthineers high-sensitivity troponin I assays (TnIH) for the Atellica IM and ADVIA Centaur XP/XPT in vitro diagnostic analyzers. The test helps in the early diagnosis of myocardial infarctions without the need for serial tropic testing. The time to first results is 10 minutes.

The most recent U.S. Food and Drug Administration (FDA) clearance was Siemens Healthineers high-sensitivity troponin I assays (TnIH) for the Atellica IM and ADVIA Centaur XP/XPT in vitro diagnostic analyzers. The test helps in the early diagnosis of myocardial infarctions without the need for serial tropic testing. The time to first results is 10 minutes. 

Feature | Cardiac Diagnostics | March 22, 2019 | Linda C. Rogers, Ph.D.
Troponins are a family of proteins found in skeletal and heart (cardiac) muscle fibers that produce muscular contract
ACC/AHA Update Guidance for Preventing Heart Disease; Stroke
Feature | Cardiac Diagnostics | March 18, 2019
The choices we make every day can have a lasting effect on our heart and vascular health. Adopting a heart healthy...
Overlay Init