A paralysed woman has been able to feed herself chocolate and move everyday items using a robotic arm directly controlled by thought, showing a level of agility and control approaching that of a human limb.
Jan Scheuermann, 53, from Pittsburgh, was diagnosed with a degenerative brain disorder 13 years ago and is paralysed from the neck down.
"It's so cool," said Scheuermann during a news conference. "I'm moving things. I have not moved things for about 10 years ... It's not a matter of thinking which direction any more, it's just a matter of thinking, 'I want to do that'."
She was shown feeding herself string cheese and chocolate unaided as well as moving a series of objects in tests designed for recovering stroke victims. She was able to do it with speeds comparable to the able bodied.
Experts are calling it a remarkable step forward for prosthetics controlled directly by the brain. Other systems have already allowed paralysed patients to type or write in freehand simply by thinking about the letters they want.
In the past month, researchers in Switzerland also used electrodes implanted directly on the retina to enable a blind patient to read.
The development of brain-machine interfaces is moving quickly and scientists predict the technology could eventually be used to bypass nerve damage and re-awaken a person's own paralysed muscles.
In the meantime, they say, systems like the one developed by the US researchers could be paired with robotic "exoskeletons" that allow paraplegics and quadriplegics to walk.
For Scheuermann, the experience has already been transforming.
"It's given her a renewed purpose," Michael Boninger, who worked on the study published in The Lancet, told Reuters. "On the first day that we had her move the arm, there was this amazing smile of joy. She could think about moving her wrist and something happened."
The research team from the University of Pittsburgh Medical Centre implanted two micro-electrode devices into the woman's left motor cortex, the part of the brain that initiates movement.
The medics used a real-time brain-scanning technique called functional magnetic resonance imaging to find the exact part of the brain that lit up after the patient was asked to think about moving her now unresponsive arms.
The electrodes were connected to the robotic hand via a computer running a complex algorithm to translate the signals that mimics the way an unimpaired brain controls healthy limbs.
"These electrodes are remarkable devices in that they are very small," Boninger said. "You can't buy them in Radio Shack."
But Boninger said the way the algorithm operates is the main advance. Accurately translating brain signals has been one of the biggest challenges in mind-controlled prosthetics.
Sign up for CIO Asia eNewsletters.