Wearable Wrist Device Could Boost Stroke Rehab

THURSDAY, Oct. 1, 2026 (HealthDay News) — A device worn on the wrist and powered by an artificial intelligence (AI) algorithm can assess rehab progress in real time for people who've lost arm movement due to stroke, new research shows. 

Right now, such assessments can only be done during doctor visits that happen rarely and take about 30 minutes each, researchers at the University of Massachusetts Amherst noted. 

The new wrist device appeared to outperform doctors in assessing how effective a particular rehabilitation program is in helping patients regain arm function.

“We are the first group to actually show that, using wearable data, we can extract information about patients’ motor severity, which clinicians can actually use to determine whether their intervention is effective or not,” said study co-author Sunghoon Ivan Lee. He's an associate professor of information and computer science at the university.

Arm disability affects more than three-quarters of the nearly 800,000 Americans who suffer a stroke each year, according to background information in a UMass Amherst news release. 

For about 40% of those patients, arm disability continues and rehabilitation is required. 

Physical therapy can work wonders in restoring movement, but doctors typically only check in on its effectiveness in two 30-minute sessions given before and after rehabilitation, the research team explained.

The new wrist device could change all that. 

It contains an accelerometer that records any arm movement. That movement is then interpreted by a machine-learning algorithm developed by Lee and a colleague. 

As Lee explained it, the severity of motor impairment a patient is experiencing is not wholly reflected in their ability to move the arm, and the algorithm accounts for that.

The less severe the impairment, "the more likely you’re going to move a lot, but they’re not exactly the same,” Lee said. “Increasing the use of the limbs — yes, we can encourage the person to make use of the limb more. But patients cannot make instant changes to motor severity through short-term behavior change.”

The model used in the device to assess rehab effectiveness is based on accelerometer readings over time, plus clinician assessment scores gathered between one week to six months after the stroke. 

Lee's team estimates that the device is 40% to 50% more accurate than a clinician's evaluation at assessing the patients' true condition during rehab. 

As it stands now, during the therapy process, "neither the patient nor the therapist has a clear idea of how patients are responding to the treatments that they’re receiving,” Lee noted. “Currently, clinicians aren’t able to see if patients are responding to the prescribed exercises, and patients have no way of knowing how they are progressing.”

The device should change that, Lee said, giving patients and their physicians daily feedback on how effective the patient's rehab is and what changes might be necessary.  

While physician assessments are rare "snapshots" in time, the device shows ongoing performance. 

Lee's team thinks checking in with the device can also help patients feel more invested in their progress, boosting their motivation to stick with therapy. 

The findings were published Sept 30 in Science Translational Medicine. 

More information

There's more on post-stroke rehabilitation at the Mayo Clinic.

SOURCE: University of Massachusetts Amherst, news release, Sept 30, 2026