The heart rate sensors in your Apple Watch, Android Wear, Garmin or Fitbit can detect early signs of diabetes with 85 percent accuracy, according to a study from app developer Cardiogram and the University of California, San Francisco.
In a study involving 14,000 Apple Watch and Android Wear users, health sensor data was used to train a deep neural network called DeepHeart to identify participants with and without diabetes. DeepHeart has previously shown high accuracy in detecting atrial fibrillation, hypertension and sleep apnea.
The 2015 Framingham Heart Study found that low heart rate variability and high resting heart rate are predictors of who will develop diabetes over 12 years. That helped inform the Cardiogram study's use of 200 million heart rate and step count measurements.
More than 100 million adults in the US have diabetes or prediabetes, according to the Centers for Disease Control and Prevention. A quarter of people with diabetes are undiagnosed, and more than 88 percent of people with prediabetes don't know they have it. Cardiogram's study demonstrates that the heart rate sensors people are already wearing -- paired with an AI-based algorithm -- could become critical tools in detecting early signs of diabetes.
The study's results come amid earlier speculation that Apple is working on years away from becoming a reality. The tech giant , which can identify irregular heart rhythms and notify users who might be experiencing atrial fibrillation., but subsequent reports stated the technology is
The Smartest Stuff: Innovators are thinking up new ways to make you, and the things around you, smarter.
Tech Enabled: CNET chronicles tech's role in providing new kinds of accessibility.