MD+DI Online is part of the Informa Markets Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

AI-Based Stroke Rehab System Shown to Be as Effective as One-on-One Therapy

AI-Based Stroke Rehab System Shown to Be as Effective as One-on-One Therapy
Image of the Motus Nova Hand Mentor and Foot Mentor courtesy of Motus Nova
Motus Nova’s “theratainment” device allows patients to perform therapeutic exercises to help make the tedious process of neurological rehab engaging and fun, while eliminating barriers to traditional physical therapy such as cost, time, and lack of motivation.

The idea that the brain is capable of rewiring after a stroke is not new, but the therapy needed to achieve this is cumbersome at best. “Studies show that neuroplasticity requires hundreds to thousands of hours of repetition of focused, concentrated training to get the brain to have those functional changes,” said David Wu, Motus Nova’s CEO, in an interview with MD+DI. But he explained that those many hours of therapy with a clinician are not always feasible for patients, because of concerns about time and logistics, cost, and sometimes even a lack of motivation.

So, Motus Nova has developed a device that can be used at home and is as appealing as playing video games. The Motus Nova Hand Mentor and Foot Mentor can be used independently and without supervision. It uses artificial intelligence (AI) to do an initial assessment for a baseline measurement, and as the user continues operating the device, it uses a suite of sensors to monitor in real time to determine what the best therapy is most effective for recovery. “We have essentially machine learning and AI in place where the more the user uses it, the better the device gets at personalizing the therapy,” Wu continued.

After testing the initial therapy device in several clinical trials, the company came up with the idea of making it more interesting by attaching it to a screen. The Hand Mentor and Foot Mentor are sleeve-like robots that fit over the patient’s hand or foot. They are equipped with an active-assist air muscle and a suite of sensors and accelerometers, which provide clinically appropriate assistance and resistance while patients perform the therapeutic exercises. A touchscreen control provides goal-directed biofeedback through interactive games.

“We call that “theratainment,” where we trick the user into doing their therapy with their stroke-affected side,” said Wu. “By controlling a spaceship or controlling a character up and down, they're avoiding obstacles or shooting enemies. To succeed at these games, they have to use their limbs. And by doing that they're making new [brain] wiring possible.”

The device collects all movement data, which they can retrieve from the device or through a computer. “We’ve found from comments from users that this is very motivational,” said Wu. “[The device] can show you since you started that, for example, you've increased your range of motion by a certain percent, or you've increased your strength by so much this week,” said Wu. He explained that the data is available for the patient to share with any number of clinicians to guide further in-person treatment sessions.

“We really see the Hand Mentor and Foot Mentor being an individual and personalized device that goes into the home, as the first step into personalized medicine that's driven by AI,” said Wu. “I think, in that regard currently, we’re just matching the effectiveness of physical therapy, but we do hope that one day with all of this evidence-based data that we can provide a faster path to recovery that's personalized for each individual, and in that way, we can really add to the overall body of knowledge for medicine,” he concluded.

TAGS: News
Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.