13 JULY 2006
Science fiction becoming medical fact?
A spinal-cord injury in your neck can lock you inside your head. Unable to move your arms or legs, you may even need help breathing. Although quadriplegics can gain a bit of control over their environment with devices that measure their breath, eye movement, or brain waves, these devices are crude at best.
Courtesy John Donoghue, Brown University.
This week, Nature is publishing an article describing a higher degree of control from electrodes implanted inside the brain. The first volunteer, a man whose spinal cord had been severed in the neck, had been injured by a knife, but similar disability can result from strokes, accidents and disease.
During a nine-month experiment, the 25-year-old learned to control a cursor on a computer screen and to open simulated email on a test program. He did not need to learn anything new, unlike users of other technology for bypassing a broken spinal cord; he simply imagined that he was moving his arm.
The test showed that the primary motor cortex in the brain, which controls movement, can make signals three years after a devastating spinal-cord injury, and that these signals could be picked up, sent outside the brain, and converted into electronic commands for a computer cursor or robot.
"We found that cortical activity can be modulated voluntarily even years after spinal cord injury," said Leigh Hochberg, lead author of the article. Hochberg is a neuroscience researcher at Brown University and a neurologist at Massachusetts General Hospital in Boston. "Some researchers might have predicted that this part of the brain would alter its function dramatically after the spinal cord was injured. But ... the movement-related signals are still there."
Donoghue is a professor of neuroscience at Brown University who has long studied how the brain initiates movement. "The project started out as an effort to understand the brain mechanism of going from thoughts or concepts of movement, and turning that into an action," Donoghue said. "We were trying to understand both, and develop a technology to examine the brain's movement control system, and understand the processing as a neuroscience question. Once we had a sensor that looked like it could be long-term implantable, and understood how the brain conceived of the movement of a hand, we could translate it for a human patient whose brain was functioning well, but who couldn't move due to damage to the spinal cord."
Most previous efforts to control movement by detecting brain waves relied on sensors located on the scalp rather than inside the brain. Scalp-based sensors are safer -- no surgery is needed -- but they detect the averaged signals of millions of neurons that usually are not related specifically to movement. As a result, patients need many hours or weeks of training before they can emit a "readable" signal using brain waves.
The strong, sensitive type
In contrast, the 96 implanted electrodes worked on the first test. When the subject imagined making a movement, such as moving the hands apart or together, he initiated a specific pattern of neuronal firing, which was detected, analyzed, and converted into movement on a computer screen or by a robot of some sort. "The first time, once we began to collect data, we asked the participant to imagine moving the hand to left, and we saw one neural pattern," says Hochberg. "We saw a different pattern when he was asked to move it to the right." "There was no learning involved in generating the signals," Hochberg told us.
Courtesy John Donoghue, Brown University
Since the ultimate goal is to restore some function to people trapped inside non-responsive bodies, the researchers also tested whether the first volunteer could control a television (yes), play a simple computer game, (yes), draw with a paint program (yes), and operate a prosthetic hand (yes). The results "suggest that, with further work, it may be possible to provide even greater environmental control, perhaps even allowing someone to eat a meal at their own pace, with only limited assistance from others," Hochberg said.
A most helpful detour
But while the study represents a major advance, the new prosthesis is a research project that is years away from the market. Although miniaturization is under way, the equipment is still clunky. And some of the implant signals suddenly degraded after a few months. Although that "seems to be the result of technology rather than biology," Hochberg said, it still needs correction.
Indeed, improvements in the technology are inevitable. For example, similar roadblocks have already been overcome with cochlear implants, which bypass broken nerves in the ear and restore some hearing to many deaf people. More than 100,000 people receive signals in the brain from cochlear or other implants, Donoghue said, but nothing on the market takes signals from the brain like the device being tested.
Although the brain is a learning machine, little learning is needed to operate the new prosthesis, Donoghue says. "What is also encouraging is the immediate response from the brain. When asked to 'think right' or 'think left,' patients were able to change their neural activity immediately. And their use of the device is seemingly easy. Patients can control the computer cursor and carry on a conversation at the same time, just as we can simultaneously talk and use our computers."
Most of us take these abilities for granted. To a quadriplegic, they are priceless.
— David Tenenbaum
• Neuronal ensemble control of prosthetic devices by a human with tetraplegia, Leigh R. Hochberg et al, Nature, 13 July, 2006.
• Interested in joining a trial?