Researchers Use Computers and ‘Exoskeletons’ to Help Stroke Survivors

Stroke survivor Ken Allsford focused intensely on how he wanted to bend his elbow.

And then the robot exoskeleton attached to his left arm obeyed his unspoken command, moving his crippled limb.

“It was a combination of exciting and trepidation, because sometimes nothing would happen,” Allsford, 61, of Katy, Texas, recalled. “But when you actually see it move without actually making the moves yourself, that’s very exciting.”

The experiment with Allsford was part of an ongoing project to see if such a brain-machine interface can help improve the rehabilitation of stroke patients.

Ten stroke patients had clinically significant improvements in their arm movement after more than a dozen therapy sessions with a robot exoskeleton powered by their own brains, researchers reported recently in the journal NeuroImage: Clinical.

“We found that there was an improvement in 80% of the participants,” said senior researcher Jose Contreras-Vidal, director of the Non-Invasive Brain Machine Interface Systems Laboratory at the University of Houston.

Most patients retained their improved function for at least two months after therapy ended, suggesting the potential for long-lasting gains, he added.

Physical therapists often help stroke patients overcome paralysis by manually moving the person’s limbs again and again, with the hope that the brain will rewire itself to restore control over the arm or leg, Contreras-Vidal said.

This type of therapy has started to rely on robots, which can effortlessly move arms or legs in specific motions without growing tired, but the use of robots has had mixed results, he said.

“A pitfall of this robotic therapy has been that they fail to engage the user. To the extent that the robot now is moving your limb or your leg, you’re not really into that. You’re moved passively,” Contreras-Vidal said. Such passive movement doesn’t help your brain restore command over paralyzed limbs.

David Putrino, director of rehabilitation innovation for Mount Sinai Health System in New York City, agreed that mindless robotic repetitions aren’t of great benefit.

“We want patients to do hundreds of repetitions of movement, but we need them to be mindful repetitions of movement, not just mindlessly performing a movement,” Putrino said. “It’s thinking about every aspect of the movement you’re performing, and maintaining a high level of engagement for every repetition.”

Contreras-Vidal and his team created a new device that would capture a person’s intention to move via their brainwaves, and then transmit that intention to a therapy robot.

An electroencephalogram (EEG) skullcap strapped to the heads of Allsford and his fellow participants monitored the parts of their brain that control movement.

The patients were asked to first think about flexing and extending their elbow, and then to gently attempt the movement, the study report explained.

When the EEG captured brainwaves signaling the intent to move, it triggered the exoskeleton to move the elbow in the manner requested by the therapist.

This required patients to remain engaged in the therapy, which researchers suspected would force the brain to adapt and improve their outcomes — and it did, for most.

“There are some people who’ve had remarkable improvement. Others did not improve as much, so we’re still at the stage of understanding what type of person with a stroke will respond best to this type of therapy,” said co-researcher Dr. Gerard Francisco. He is chairman of physical medicine and rehabilitation at McGovern Medical School at The University of Texas Health Science Center at Houston.

“My suspicion is that those who have some movement or who have more movement before they start this protocol are more likely to recover more, but I could be wrong,” he said.

Each therapy session lasted about three hours, and people took part in 14 to 15 sessions.

This was a pilot study, and Francisco said researchers plan to test their findings with more people.

Putrino hopes future studies will also include people receiving standard physical therapy, either with humans or robots, so researchers can compare results against patients using the brain-machine interface.

“The big burden of proof in this research now is we need to know if a brain-powered robot outperforms robotics alone,” Putrino said. “That’s the question for the field, and they did not have that control group in their cohort, which is a real shame.”

More information

The American Stroke Association has more about stroke rehabilitation.

SOURCES: Ken Allsford, Katy, Texas; Jose Contreras-Vidal, PhD, director, Non-Invasive Brain Machine Interface Systems Laboratory, University of Houston; David Putrino, PhD, director, rehabilitation innovation, Mount Sinai Health System, New York City; Gerard Francisco, MD, chairman, physical medicine and rehabilitation, McGovern Medical School, University of Texas Health Science Center at Houston; NeuroImage: Clinical, Nov. 19, 2020