Machine learning identifies diabetics with 85% accuracy using wearable heart rate monitors

Adjust Comment Print

With that information, Cardiogram was able to develop parameters for detecting the signs of diabetes, and train its deep learning network, DeepHeart, on those parameters.

The latest study - a collaboration between University of California San Francisco (USCF) and Apple Watch app-maker Cardiogram - is the latest to push the health benefits of the popular smartwatch, which shifted 18 million units in the last three months of 2017 alone.

In a new study conducted with the UCSF Department of Medicine, a neural network developed by a startup called Cardiogram was able to detect diabetes with almost 85 percent accuracy, just by looking at people's heart beats over time. Although the app is available through Apple Watch or Android smartwear, the study only involved Apple Watch users. This is done via the heart-rate monitor, or to be more specific, the light emitted by the monitor which when combined with neural networks is said to be able to detect diabetes in the user with an 85% accuracy rating.

While the app isn't created to diagnose prediabetes or diabetes, it would alert a user to check in with his or her healthcare provider.

"Typical deep learning algorithms are data-hungry, requiring millions of labeled examples, but in medicine, each label represents a human life at risk-for example, a person who recently suffered a heart attack or experienced an abnormal heart rhythm", Hsieh explained in a Q&A. Better awareness of heart health, diet, and exercise-all of which the Apple Watch can help with-could help wearers in that period to improve their lifestyle and either stall or reverse the onset of this disease. Once the program was "trained" to detect diabetes, a separate set of data was used to establish the program's accuracy in identifying which of the participants were diabetics.

Positive Thinking about Aging May Protect Against Dementia, Study Finds
Stress may be a key factor in the development of dementia, the Australian Broadcasting Corporation (ABC) reported . It is not clear, why the rest of the persons with this genetic makeup do not develop dementia.

"Twenty-four percent of people with diabetes, and 88.4 percent with pre-diabetes, don't realize they have it", Brandon Ballinger, CEO of Cardiogram, told MobiHealthNews in an email.

"Your heart is connected with your pancreas via the autonomic nervous system", Cardiogram co-founder Johnson Hsieh pointed out in the release. In addition, a 2015 Framingham Heart Study showed that high resting heart rate and low heart rate variability predicts who will develop diabetes over a 12-year period. The app was also able to detect high blood pressure with 80 percent accuracy, and sleep apnea with 83 percent accuracy. According to Cardiogram, DeepHeart has been used in past studies to detect atrial fibrillation, hypertension and sleep apnea.

Cardiogram's work could eventually prove useful in acting as a pre-screen for these kind of conditions.

Going forward, Cardiogram is planning to incorporate DeepHeart directly into the Cardiogram app to help in detection of the early signs of disease.