Natural sense of touch restored with bionic hand

We have the world at our fingertips. A sense of touch can sometimes be as important as sight, helping us to avoid crushing delicate objects or ensuring that we hold on firmly when carrying hot cups of coffee. Now, for the first time, a person who lost his left hand has had a near-natural sense of touch restored thanks to a prosthesis.

"I didn't realise it was possible," says Dennis Aabo Sørensen, who is so far the only person to have been fitted with the new prosthesis. "The feeling is very close to the sensation you get when you touch things with your normal hand."

To restore Sørensen's sense of touch, Silvestro Micera at the Swiss Federal Institute of Technology in Lausanne and his colleagues implanted tiny electrodes inside the ulnar and median nerve bundles in Sørensen's upper arm. Between them, the ulnar nerve – which runs down to the little finger and ring finger – and the median nerve – which runs down to the index and middle fingers – carry sensations from most of the hand, including the palm.

The team then connected the electrodes to pressure sensors on the fingertips and palm of a robotic prosthetic hand via cables running down the outside of Sørensen's arm. When he used the hand to grasp an object, electrical signals from the pressure pads were fired directly into the nerves, providing him with a sense of touch.

Getting to grips

The electrical signals were calibrated so that Sørensen could feel a range of sensation, from the slightest touch to firm pressure just below his pain threshold, depending on the strength of his grip.

According to Sørensen, the feelings were very natural and close to those in his real hand. So much so, in fact, that he can differentiate objects held in his prosthetic hand by touch alone. For example, he could tell the difference between cotton wool, a plastic cup and a block of wood about 90 per cent of the time while wearing a blindfold.

Knowing how firmly he was grasping an object was a game-changer. "With my existing prosthesis I always have to look at what I'm doing with it," says Sørensen. Without a sense of touch to guide him, picking up a piece of fruit can end messily, he says. "It's an amazing product, I'd take one tomorrow if I could."

However, regulations governing the use of surgically implanted electrodes mean that Sørensen had to give up the prosthesis after 30 days. Before the treatment becomes clinically available, Micera's team needs to test its long-term viability.

One challenge is to ensure that the implanted electrodes continue working long after they have been inserted. For a first attempt, results are promising, says Micera. After 30 days, 90 per cent of the electrodes still worked.

In the meantime, Dustin Tyler at Case Western Reserve University in Cleveland, Ohio, has already shown that a slightly less invasive procedure can still be effective after 18 months. Tyler's team implanted electrode "cuffs" that clamped around nerve bundles in the arms of two people who had lost a hand, again restoring a limited sense of touch. The cuffs work in a similar way to Sørensen's implants but, because they do not enter the nerves, greater changes in electrical input are needed for stimulation. This makes the sensations coarser, says Micera.

Personal touch

Tyler agrees that stimulating individual fibres in the nerve bundles leads to finer degrees of sensation. In theory, by precisely stimulating the relevant nerve fibres it could be possible to make a prosthetic limb feel exactly like a real one – restoring not only fine variations of touch, but also sensations of hot and cold, and even pain. This will be more challenging, however, because the nerve fibres that carry these signals are very small, says Tyler.

Recent studies have shown that macaques can also regain a sense of touch by having electrical signals from a prosthesis routed directly to their brains. But going via the nerves in the arm has the advantage of using the brain's usual channels of input, allowing the signal to be interpreted more naturally. "You have the brain helping you do the job," says Micera.

Micera's approach is conceptually similar to how cochlear implants have provided hearing for tens of thousands of people, says Greg Clark at the University of Utah in Salt Lake City. Cochlear implants stimulate nerve fibres in the same way they would normally be activated by sound itself. "We don't need to know the code that the brain uses to interpret speech for this approach to work," he says. "We just need to provide the basic inputs. Then the brain interprets these inputs in much the way it normally would."

By doing the same with sensory input from skin and muscles, we are on the way to restoring much more than hearing loss. This will not only have practical benefits, says Clark, but could literally help people with amputations feel whole again.