Soft exoskeleton aims to restore hand mobility for people with motor impairments

John Nassour, Nicholas Berberich and Gordon Cheng test a soft exoskeleton for the hand.

Millions of people around the world have difficulty moving their arms due to brain, spinal cord or nerve injuries. Researchers from the Technical University of Munich and the Passauer Wolf Rehabilitation Center have developed a new lightweight soft exoskeleton that can restore the ability to move your arms and grasp objects.

How the new development works

The new robotic system, described in an article in the journal Nature Machine Intelligence, combines advanced mechanics, muscle signal reading technology and artificial intelligence. “Our recent paper looked at restoring hand function in people with severe impairments,” Gordon Cheng, who led the study, told Tech Xplore. - Hand function is critical in every aspect of life. The patient we worked with almost completely lost the mobility of his hand. Our task is to help such people.”

Unlike conventional rigid robots, the new devices are made from soft, flexible materials that better adapt to the contours of the body. But many existing hand exoskeletons have limitations. Some restore movement only in those who have partial control of the arm. Others don't allow you to deftly manipulate objects - only perform simple actions like touching or grabbing.

Exoskeleton with AI that deciphers weak muscle signals

A new exoskeleton created by Cheng and his colleagues restored hand mobility to an ALS patient who had been unable to move his right arm for several years. The device is equipped with sensors that detect the electrical activity of the hand muscles. A machine learning algorithm analyzes this activity and predicts when the user wants to grab the item. The exoskeleton then performs movements to grasp the desired object.

“The soft exoskeleton for the hand is sewn from fabric (like regular clothing) on ​​a standard sewing machine and powered by air. It works like a glove—light and safe,” Cheng explained. - The key element is a model that predicts the patient's intention to take the object. For example, a patient with severe ALS was able to consciously pick up a fork and eat a piece of cake independently - after almost four years of paralysis.”

Gordon Cheng, who led the latest study.

Clinical trials and results

Cheng and his colleagues tested their exoskeleton in a series of tests. They involved a patient with ALS and six people who, after a stroke, could not move one arm. Participants were asked to wear an exoskeleton and perform two tasks: the Box and Block Test and the ARAT (Action Research Arm Test). Both are widely used in clinics to assess hand dexterity.

The researchers found that those with the most severe disabilities benefited most from the exoskeleton. But those who had partial control of their hand did not always perform better on tasks with the device.

“The biggest lesson we learned was from an ALS patient,” says Cheng. “We worked together, and he himself made changes to the exoskeleton and control system. For example, he did not want the glove to move automatically - he insisted that he himself give the command to move. This once again proves how important the predictor of intentions is.”

Participants perform ARAT tasks to assess hand function using a soft exoskeleton.

Prospects and plans of developers

In the future, the soft robotic system can be further developed and tested in other clinics. And in the future, it will allow people who, due to injury, stroke or neurological disease, cannot move their arm, can independently perform everyday tasks.

“We are now developing an exoskeleton for the lower body - it will help more people,” adds Cheng. “My goal is to make such technologies available to as many people as possible in the coming years, until retirement.”