The Apple Watch may become a useful tool for not only for fitness, but for monitoring serious medical conditions. Cardiogram, a heartbeat measurement app, and the University of California, San Francisco teamed and found that the Apple Watch is 97% accurate in detecting the abnormal heart rhythm problems when paired with neural network algorithms.
Of the 6,158 participants—which were initially recruited through the app itself—200 had been diagnosed with paroxysmal atrial fibrillation (AF), the most common type of heart rhythm problem. Engineers used those subjects to train a deep learning system to distinguish between patients with normal and abnormal heartbeats.
Cardiogram then tested the system against 51 in-hospital procedures that restore the heart’s normal rhythm. The algorithm and Apple Watch successfully found, with 97% accuracy, irregular heart activity.
Since the algorithm is an early version that could be made even more accurate, this study holds a lot of promise. Atrial fibrillation causes one in four strokes, many of which can be prevented with inexpensive drugs. Yet, detecting one before it happens is difficult. With this wearable technology, doctors could potentially save the lives of more at-risk populations.
The app plans to do more work before using its deep learning system on other Cardiogram users. Here is Sarah Buhr, writing for TechCrunch:
In the meantime, Cardiogram and UCSF will continue its eHealth study and plan to further validate its deep neural network “against multiple gold standards, incorporating the results into the Cardiogram app itself, and investigating the ability to detect health conditions beyond atrial fibrillation,” according to Cardiogram.
Once they perfect the algorithm, it could become another example of how technology has evolved to help save lives.