Computer interface links signals from the brain to an artificial limb, giving the wearer better balance, flexibility and speed

Bionic leg moves like a natural limb — without conscious thought

Computer interface links signals from the brain to an artificial limb, giving the wearer better balance, flexibility and speed

A robotic leg that can be fully controlled by the brain and spinal cord has enabled seven people who had lost a lower leg to walk roughly as fast as people without amputations.

The bionic limb uses a computer interface that amplifies nerve signals from muscles in the remaining part of the leg and allows the wearer to move the prosthesis with their own thoughts and natural reflexes.

In a clinical trial involving 14 people, participants with this interface were able to walk 41% faster than were those with standard robotic legs. They also had better balance and ability to change their speed, climb stairs and step over obstacles. The results were published today in Nature Medicine¹.

“This is the first study that demonstrates natural gait patterns with a full neural modulation where the person’s brain is 100% in command of the bionic prosthesis, not a robotic algorithm,” said study co-author Hugh Herr, a biophysicist at the Massachusetts Institute of Technology in Cambridge, at a press conference announcing the findings.

“Even though the limb is made of titanium and silicone and all these various electromechanical components, the limb feels natural, and it moves naturally without even conscious thought,” he added.

Herr had both of his legs amputated after being caught in a blizzard while ice climbing on New Hampshire’s Mount Washington in 1982. He says he would consider using the interface devices for his limbs in future.

Muscle meets machine

Most existing bionic artificial limbs rely on preset algorithms to drive movement and can automatically switch between predefined modes for various walking conditions. Advanced models have helped people with amputations to walk, run and climb stairs more fluently, but the robot, rather than the user, retains control of the leg movement, and the device doesn’t feel like part of the body.

Determined to change this, Hurr and his colleagues developed an interface that controls the robotic limb with signals from the nerves and muscles that remain after amputation.

Their clinical trial included 14 participants with below-knee amputations. Before wearing the robotic device, seven of them underwent surgery to link together pairs of muscles in the residual sections of their legs..

This surgical technique, which creates what’s called an agonist–antagonist myoneural interface (AMI), aims to recreate natural muscle movements so that the contraction of one muscle stretches another. It helps to reduce pain, preserve muscle mass and improve comfort with the bionic limb².

The bionic leg itself includes a prosthetic ankle embedded with sensors, along with electrodes that are attached to the surface of the skin. These capture electrical signals produced by the muscles at the amputation site and send them to a small computer to be decoded. The leg weighs 2.75 kilograms, similar to the average weight of a natural lower limb.

Fast improvements

To test the system, the participants practised using their new bionic legs for a total of six hours each. Then, the researchers compared their performance on various tasks with that of the seven other participants, who had received conventional surgery and prostheses.

The AMI increased the rate of muscle signals to an average of 10.5 impulses per second, compared with around 0.7 impulses per second in the control group. Although this is equivalent to only 18% of muscle signals in biologically intact muscles — which is around 60 impulses per second — participants with the AMI were able to fully control their prostheses and walked 41% faster than those in the control group. Their peak speeds matched those of people without amputations when walking on flat ground along a 10-metre-long hallway.

“I actually found it remarkable that with so little learning, they were able to achieve such good results,” says Levi Hargrove, a neural engineer at Northwestern University in Chicago, Illinois. “They would see even more benefit with a longer accommodation period, wearing the device.”

The researchers also tested how well the participants could navigate various situations, including walking on a surface with a 5-degree slope, climbing stairs and stepping over obstacles. In all scenarios, AMI users showed better balance and faster performance than did people in the control group.

“It gives the user such a high flexibility that is much closer to how the biological leg works,” says Tommaso Lenzi, a biomedical engineer at the University of Utah in Salt Lake City.

Natural experience

The technology offers new hope to people with amputations who want to regain a natural walking experience. “People who have an amputation want to feel in control of their limbs. They want to feel the limb to be part of their body,” says Lenzi. “This kind of neural interface is necessary to create that.”

Improvements in the leg’s design could include making it lighter and upgrading the surface electrodes, which are sensitive to humidity and sweat and might not be suitable for everyday use, says Lenzi. And future studies will be needed to test whether the device can handle more demanding activities, such as sprinting and jumping.

Herr says his team is already looking for ways to replace the surface electrodes with small implanted magnetic spheres that can accurately track muscle movements.

This trial “provides the foundation that we need to then go and translate this into clinically viable technologies and solutions for everyone with an amputation”, says Lenzi.

doi: https://doi.org/10.1038/d41586-024-02157-3

This story originally appeared on: Nature - Author:Miryam Naddaf