Amputees often feel disconnected from their bionic hands. AI could bridge the gap : Shots

Researchers have created a prosthetic arm that, with the help of artificial intelligence, can act like a natural one.

The key is for the hand to recognize when the user wants to do something and then share control of the movements needed to complete the task.

An approach combining artificial intelligence with special sensors helped four people who were missing an arm simulate drinking from a cup, scientists say. Marshall Troutresearcher at the University of Utah and lead author of the study.

When the sensors and artificial intelligence assisted, participants could “very reliably” pick up a cup and pretend to take a sip, Trout said. But without that overall control of the bionic arm, he said, they “would squash or drop it every time.”

Success, described in the magazine Natural communicationsis notable because “the ability to apply grip force is one of the things we're really struggling with in prosthetics right now,” says John Downeyassistant professor at the University of Chicago who was not involved in the study.

According to him, such problems lead to the fact that many amputees become disillusioned with their bionic arms and stop using them.

Helping hand

The latest bionic hands are equipped with motors that allow them to rotate, move individual fingers and manipulate objects. They can also detect electrical signals coming from the muscles that are used to control these actions.

But as bionic arms have become more capable, they have become more difficult for users to control, Trout says.

“The person has to sit and really concentrate on what they're doing,” he says, “which isn't really the way an uninjured hand behaves.”

For example, the natural hand requires very little cognitive effort to perform routine tasks such as reaching for an object or tying a shoelace. This is because once a person puts a task into action, most of the work is done by specialized circuits in the brain and spine that take over control.

These circuits allow you to perform many tasks efficiently and automatically. Our consciousness intervenes only if, say, a lace breaks or an object moves unexpectedly.

So, Trout and a team of scientists decided to create a smart prosthesis that would more closely resemble a person's own hand.

“I just know where my coffee cup is and my hand naturally grips and touches it,” he says. “That’s what we wanted to recreate with this system.”

The team turned to artificial intelligence to take over some of these subconscious functions. This meant detecting not only the signal coming from the muscle, but also the intention behind it.

For example, the AI ​​control system has learned to detect the slightest twitch of the muscle that flexes the arm.

“That's when the machine's controller kicks in and says, 'Oh, I'm trying to catch something, I'm not just sitting still,'” Trout says.

To make this approach work, the scientists modified the bionic arm by adding proximity and pressure sensors. This allows the artificial intelligence system to measure the distance to an object and estimate its shape.

Meanwhile, pressure sensors on the fingertips tell the user how firmly their prosthetic hand is gripping an object.

Shared governance

The idea of ​​shared control of a bionic arm is aimed at how many people react when they use a prosthetic with superhuman abilities, he says. Jacob GeorgeProfessor at the University of Utah and Director of the Utah Neurorobotics Laboratory.

“You can create a robotic arm that can perform tasks better than a human user,” he says. “But when you actually give it to someone, they don't like it.”

According to him, this is because the device seems foreign to them and out of their control.

John Downey says that one of the reasons we feel connected to our hands is because they are controlled jointly by our thoughts and reflexes in the brain stem and spinal cord.

This means that the thinking part of our brain doesn't have to worry about the details of every movement.

“All of our motor functions involve subconscious reflexes,” says Downey, “so providing robotic imitation of these reflex loops will be important.”

George says a smart bionic arm solves this problem.

“A machine does something, and a person does something, and we combine the two,” he says.

This is an important step towards creating prosthetic limbs that feel like an extension of a person's own body.

“Ultimately, when you create an embodied robotic arm, it becomes part of the user's experience, part of who they are, rather than just a tool,” says George.

According to Downey, even the most advanced bionic arms still require the help of the human brain.

For example, a person can use the same natural hand to carefully thread a needle and then firmly lift the child.

“The dynamic range is far beyond what robots can typically handle,” says Downey.

This is likely to change as bionic limbs become more versatile and capable. What won't change, scientists say, is people's desire to maintain a sense of control over their artificial appendages.