Thursday, February 8, 2018

Using heart rate data (and machine learning) to detect diabetes?

A very interesting study seems to suggest that continuous heart data could identify patients who have diabetes. You can read more about this story on Wired:
... at the annual AAAI Conference on Artificial Intelligence in New Orleans, digital health-tracking startup Cardiogram presented research suggesting the Apple Watch’s heart rate sensor and step counter can make a good guess at whether or not a person has diabetes—when paired with the right machine-learning algorithms, of course.
In 2013, researchers at UCSF launched the Health eHeart study and registered close to 200,000 participants. About 40,000 opted to link their health information with their Cardiogram app. The DeepHeart neural network was trained to spot patterns and trends linked to human disease. Using semi-supervised sequence learning (artificial intelligence), the machine interpreted patterns of heart rate variability and was able to identify patients with diabetes 85% of the time.

I find myself wondering if some of this may be related to certain pharmacologic agents such as beta blockers...

No comments:

Post a Comment

Blog Widget by LinkWithin