Withings' rumored heart rate monitor, AI to predict cognitive decline, and more digital health news briefs

Also: Biodegradable implant accelerates nerve damage repair; algorithm uses wearables to predict behavior, blood pressure changes.
By Dave Muoio and Laura Lovett
Share

A peak at Withings' next device? Rumor has it Withings is creating a new fitness tracker at a lower price point. Wearable reports  that a tracker listed by the company as Pulse HR in a recent FCC filing, which could indicate that it will have a heart rate monitor.

For the last two year Withings has been part of Nokia’s digital health division. But the department faced challenges from the get go and in May Eric Carreel, cofounder of the original Withings, bought back Nokia’s digital health division two years after it was sold to the Finnish tech giant. At the time of the purchase Carreel said he planned to relaunch the Withings brand by the end of 2018.

Devices and predictive AI work well together. Thanks to machine learning, wearables and other off-the-shelf technology can be used to not only measure patients’ blood pressures and behaviors but also predict changes in each, according to a recent investigation. In it, researchers from the University of California, San Diego outfitted eight patients with FitBit Charge HR and an Omron Evolv wireless blood pressure monitor and collected data on their sleep, exercise, and blood pressure. With these, they were able to build an algorithm that made daily predictions regarding these health metrics, offered effective suggestions, and resulted in substantially healthier behaviors and blood pressure for the individual users.

"This research shows that using wireless wearables and other devices to collect and analyze personal data can help transition patients from reactive to continuous care," Sujir Day, the paper’s coauthor and the director of the UCSD School of Engineering’s Center for Wireless Communications, said in a statement. "Instead of saying 'My blood pressure is high therefore I'll go to the doctor to get medicine,' giving patients and doctors access to this type of system can allow them to manage their symptoms on a continuous basis."

Machine learning to identify decline. Research recently published last month in PLOS Computational Biology offers an example of how machine learning technology could be used to predict cognitive decline and subsequent Alzheimer’s disease. By training an algorithm with a longitudinal dataset of healthy and declining seniors and then validating it against another independent body of data, researchers found that their tool was reasonably capable of predicting decline.

"At the moment, there are limited ways to treat Alzheimer's and the best evidence we have is for prevention,” Dr. Mallar Chakravarty, assistant professor at the department of psychiatry at McGill University, said in a statement. “Our AI methodology could have significant implications as a 'doctor's assistant' that would help stream people onto the right pathway for treatment. For example, one could even initiate lifestyle changes that may delay the beginning stages of Alzheimer's or even prevent it altogether.”

Insert and forget. A collaborative team of Northwestern University engineers and Washington University School of Medicine neurosurgeons have developed a biodegradable, wireless implant that delivers pulses of energy directly to damaged nerves to accelerate their recovery. The paper-thin device, which was detailed this week in Nature Medicine, lasts roughly two weeks before dissolving, and was shown to improve recovery in rats.

"We know that electrical stimulation during surgery helps, but once the surgery is over, the window for intervening is closed," Dr. Wilson "Zack" Ray, a senior author of the research article and an associate professor of neurosurgery, biomedical engineering and of orthopedic surgery at Washington University, said in a statement. "With this device, we've shown that electrical stimulation given on a scheduled basis can further enhance nerve recovery.”

Digital physician assistant picking up users. AI startup Suki announced today that its voice-enabled digital physician assistant tech is being employed in roughly 1,000 patient encounters each week in eight states since launching in early May. By responding to voice commands and automatically updating a patient’s EHR, the company said that its platform so far has led to a 70 percent overall reduction in the time physicians are spending on updating their medical notes.

“We are proud of the progress made since introducing Suki in May, and excited to be freeing more and more physicians from the arduous burden of medical documentation to focus on their first priority – patient care,” Punit Soni, CEO and cofounder of Suki, said in a statement. “This growth demonstrates the demand for new solutions that support physicians throughout their care delivery, and work across specialties and settings.”