Building on two decades of work, neuroscientists are making inroads on ways to translate brain signals into messages to help communicate with paralyzed patients who are "locked in," unable to move any muscles.
Some of these paralyzed patients have been forced to rely on eye blinks or facial muscle twitches that are interpreted by devices on computers to slowly send messages.
In the journal Current Biology, a neuroscience team led by Bettina Sorger of Maastricht University in the Netherlands reports the success of MRI scanning machines in translating signals in the brains of six healthy volunteers into letters to enable them to spell messages. The approach represents an alternative to measuring brain wave signals from scalp electrodes, a technology now being used in a different trial involving famed physicist Stephen Hawking, who is paralyzed with ALS, also known as Lou Gehrig's disease.
"This is not mind-reading. It is under the person's control, so it is more like a typewriter for the brain," Sorger says. In the study, the volunteers used their brains to imagine actions -- perhaps drawing a boat to represent the letter "b" -- that were translated into letters by the scanners and finally into whole messages. It typically took an hour or less for volunteers to learn how to do it.
The advance comes one month after Brown University researchers reported that two "locked in" patients had manipulated a robotic arm with signals from brain implants. Neurobiologist Andrew Schwartz of the University of Pittsburgh, who was not part of that team, is expected to report a similar experiment on different patients this summer. He pointed to the efforts as a sign of rapid advances in neuroscience.
The work in the Netherlands involved MRI machines, massive and expensive medical-imaging devices that are commonly found in hospitals, where they are used to analyze injuries to organs, joints and other soft tissues. In this case, the MRI machines picked up signals of increased blood flow in precise regions of the brain as volunteers thought of tasks that represented certain letters. The scanner decoded signals using a "data-mining" computer technique essential to translating brain signals into words. "For affected patients, time is not the most pressing matter, compared to just being able to communicate," Sorger says.
"The method, along with the state-of-the-art data-mining technique, can be used to create a brain-controlled keyboard," says Harvard neuroscientist Seung-Schik Yoo, who was not part of the study. He suggests the MRI machine-based method eventually will serve as a "trainer" for paralyzed patients, who will then use less bulky, less expensive devices that detect electrical signals from the scalp tuned to their precise brain signals. Already in use, that method is faster but has not worked very well for the most severely "locked in" patients, Sorger says.
About 260,000 U.S. patients have spinal cord injuries, which, combined with traumatic brain injuries and strokes, cause roughly 1 million disability cases a year, the American Heart Association estimates. Patients are "locked in" to their brain, left without movement or speech, only in the most severe cases or in progressive diseases such as ALS.
"For me, if this helps even one person, it will have been worth the effort," says Sorger, who first came up with the way to translate MRI signals into letters as she was drifting off to sleep.
"I wrote the idea down in a hurry.I didn't want to lose it."
To see more of USAToday.com, or to subscribe, go to http://www.usatoday.com
Copyright 2012 USA TODAY, a division of Gannett Co. Inc.