Apple Watch Can Detect Abnormal Heart Rhythms
The eHeart study at the University of California, San Francisco has shown that a “deep neural network” fed data from the Apple Watch app Cardiogram was 97 percent accurate in detecting the most common type of abnormal heart rhythm, paroxysmal atrial fibrillation. Paroxysmal atrial fibrillation causes about 25 percent of strokes and two-thirds of those strokes are preventable with relatively inexpensive drugs, making this potentially life-saving news. Cardiogram and UCSF plan to further validate the neural network’s findings against external data and incorporate the results into the Cardiogram app itself. The team also intends to see if the system can detect health conditions beyond atrial fibrillation.