Do you still think you cannot walk after an injury?
Here’s what will leave your mouths open.
Artificial intelligence is helping people regain their mobility after certain neurological injuries. A robotic harness controlled by a neural network offers tailored treatment that has immediately improved their ability to walk normally.
To avoid persistent difficulties walking after a stroke or spinal injury, walking assistance is crucial. But this is a slow process that, if done wrong, can lead to a permanently impaired gait.
In the past, several physiotherapists were needed to physically support and guide each person through the process of learning to walk again. But staff are expensive, so robotic harnesses have recently been introduced.
While such harnesses can be personalised at the start of a treadmill walking session, they don’t adjust dynamically to correct a person’s particular gait in real time – which can actually hinder improvement.
So for example,unlike a physiotherapist, a harness can’t spot if someone is moving unnaturally and correct them. And if a person keeps repeating abnormal movements, these risk becoming a permanent part of their gait.
The new smart harness, developed by researchers from NCCR Robotics at the Swiss Federal Institute of Technology (EPFL) in Lausanne and the Lausanne University Hospital, prevents this from happening. Rather than simply supporting a person’s weight, as existing harnesses do, it can also correct their gait by pushing them forward or back, or side to side.
It does this after collecting information on leg movement, stride and muscle activity from body sensors and feeding it into an algorithm. The harness then provides support tailored to the way the person walks, deciding how much force to apply to their trunk to produce a natural gait.
The new system improved the in-harness gait of people following a stroke or a spinal injury. And after a single, 1-hour training session with the smart harness, people with spinal cord injury showed immediate improvement in their gait out of the harness over those given no physio session at all, the authors report today.
Artificial intelligence is widely used in rehabilitation, says Farshid Amirabdollahian at the University of Hertfordshire, UK. “We have multiple products in the market where robots help individuals recover from stroke,” he says. These include robotic gloves that exercise the hand and wrist, and a rehabilitation robot that encourages and aids repetitive exercises at home.
For walking assistance, even robotic harnesses that operate without algorithms have become smarter at load balancing and understanding how far to push people, using readings on muscle activity and brain scans.
“As the systems become more clever, the results of the rehabilitation become much better.”
The next goal is to commercialise the smart harness.
Journal reference: Science Translational Medicine, DOI: 10.1126/scitranslmed.aah3621