This Robotic Harness Could Help People Relearn to Walk After Injury

Swiss researchers have developed an algorithm-backed “smart” harness to help stroke and spinal cord injury victims practice walking in a more natural way.

A visualization of the harness. Courtesy EFPL

For patients learning to walk again after a stroke or spinal cord injury, the rehabilitation process can be slow and arduous. The traditional approach involves one or more therapists holding the patient up as they haltingly put one foot in front of the other. It’s tough, sweaty and labor-intensive.

Now, Swiss scientists have developed a robotic harness to help make things easier. The harness, which is attached to the ceiling, is backed by a deep neural network algorithm that can “learn” where a particular person needs the most support. Using the harness, patients who normally need heavy assistance to walk can roam freely around the therapy room.

“With this technology, we believe that we can improve the way rehabilitation is performed in order to enhance the recovery of gait and balance,” says Jean-Baptiste Mignardot, a neuroscientist at the Center for Neuroprosthetics and Brain Mind Institute at the Swiss Federal Institute of Technology who worked to develop the harness.

In an initial study, patients who used the harness were able to walk more naturally. The harness helped them with some of the elements of walking that most of us take for granted: balance, limb coordination, foot placement, steering.  For patients in the study who could already walk on their own using supports like walkers, after practicing with the harness their solo walking showed immediate improvement. The findings were published last month in the journal Science Translational Medicine.

Harnesses that support patients relearning to walk are not new. They are already commonly used in rehabilitation centers to take some of the weight off therapists. But these harnesses pull upward, making the patient shift their body weight backward. This creates an unnatural condition for walking that could potentially impact rehabilitation. The Swiss team developed a computational model that can predict the right configuration of forces to be applied to the patient’s trunk to simulate normal walking conditions. The harness learns how a particular patient tends to move and where they tend to shift their weight, and adjusts accordingly. This gives physical therapists a tool to potentially make the rehab process more efficient. In addition, the harness can be used to move in multiple dimensions rather than simply forward, which lets patients practice a variety of movements – zigzagging between obstacles, moving horizontally along the image of a ladder projected on the floor, sitting and standing.

Experts have cautioned that it could be a “long road” to having the system widely available to the public. The next step will be more and larger studies, including ones that compare the smart harness with traditional versions. Mignardot and his team members are also working with a medical technology company to commercialize a version of the harness, called RYSEN. They've submitted patents for the technology. 

Previous research has suggested that high-tech approaches are not always best when it comes to rehabilitation. A 2011 Duke University study, the largest stroke rehabilitation study ever conducted, concluded that simple, at-home physical therapy was the most successful method of stroke rehabilitation. “Locomotor training, including the use of body-weight support in stepping on a treadmill, was not shown to be superior to progressive exercise at home managed by a physical therapist,” the study authors wrote.  

It’s not yet clear whether the smart harness system will change this equation. But Mignardot hopes it will at least be an important part of the stroke and spinal cord injury therapy process in the future.

“[Now] physical therapists have a tool that helps them to tailor each session to the real needs of their patients,” he says.

Smart Walking Assistant

Get the latest stories in your inbox every weekday.