Check out all the on-demand sessions from the Intelligent Security Summit here.

Heart monitoring tools have already shrunk from the size of toasters into smartwatches, so the next step might not surprise you: South Korean researchers have successfully tested heart monitoring technology in a wearable “smart ring” backed by a deep learning algorithm, reports Cardiology Today, and they expect that consumer rings could be used to detect atrial fibrillation in the foreseeable future.

Research presented at this week’s Heart Rhythm Society Scientific Sessions event compared simultaneous ECGs and optical sensor-based photoplethysmographs taken in 119 patients with atrial fibrillation (AF). A convolutional neural network trained on photoplethysmography data was 99.3% accurate in diagnosing AF and 95.9% accurate in diagnosing a regular (sinus) rhythm, numbers that went up to 100% and 98.3% respectively when low-quality samples were filtered out. The researchers said the average confidence level was 98.6% for true classification and 80.5% for false classification.

“The diagnostic performance is comparable to medical-grade conventional pulse oximeters,” said Seoul National University Hospital assistant professor of cardiology Eue-Keun Choi, MD, PhD. “We would like to evaluate the deep learning algorithm with a newly developed ring device in daily activity. This will provide feasibility for AF screening in a high-risk population. Also, we hope that the ring device could be used for AF detection in a clinical trial due to its noninvasiveness.”

As smartwatches presently suffer from battery life limitations, commonly requiring daily or weekly recharges, the prospect of moving heart monitoring into an even smaller wearable may appear to be daunting. However, dropping many of the requirements of a watch — a screen, charger, and more sophisticated processor — could enable the heart monitoring, wireless, and battery components to fit comfortably within a much smaller package that could be even easier for some users to wear.


Intelligent Security Summit On-Demand

Learn the critical role of AI & ML in cybersecurity and industry specific case studies. Watch on-demand sessions today.

Watch Here

That’s been the key hurdle faced by prior heart monitoring devices, which were once large and extremely expensive, only recently evolving into affordable, daily-worn companions for some users. A handful of companies have already released or patented finger-based wearables, generally focused on limited audio, gesture, or door access applications.

VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact. Discover our Briefings.