A severely paralyzed man who lost his ability to speak has been given a voice again after scientists implanted a device to decode his brain waves — a potentially game-changing medical breakthrough for people who cannot communicate due to stroke, accidents or illness.
Edward Chang, a neurosurgeon who led the work at the University of California, San Francisco, said the speech neuroprosthesis enabled the man in his late 30s to communicate in sentences by translating signals from his brain to the vocal tract into text on a computer screen.
“To our knowledge, this is the first successful demonstration of direct decoding of full words from the brain activity of someone who is paralyzed and cannot speak,” Chang said in a statement Wednesday. “It shows strong promise to restore communication by tapping into the brain’s natural speech machinery.”
The device was tested on a man who had a massive brainstem stroke more than 15 years ago, damaging the link between his brain, vocal tract and limbs. He previously used a pointer attached to a baseball cap to select letters on a screen to communicate, researchers said.
Video released by UCSF shows the man, identified only as BRAVO1, communicating after the electrode device was implanted over the area of his brain that normally controls the vocal tract.
“How are you today?” researchers asked the man, whose brain activity was then translated via computer algorithm in real-time.
“I am very good,” the man’s response read.
The algorithm can differentiate among 50 words that could be used to generate more than 1,000 sentences, Chang’s team said.
“We decoded sentences from the participant’s cortical activity in real-time at a median rate of 15.2 words a minute, with a median word error rate of 25.6%,” researchers said.
“In post hoc analyses, we detected 98% of the attempts by the participant to produce individual words and we classified words with 47.1% accuracy using cortical signals that were stable throughout the 81-week study period.”
One of the lead authors of the study, which was published Thursday in the New England Journal of Medicine, said it took up to four seconds for the word to appear on the screen after the man tried to say it – which is quicker than tapping out a response but still slower than normal speech.
The device decoded words at a rate of up to 18 words per minute, with up to 93 percent accuracy, researchers said.
“We were thrilled to see the accurate decoding of a variety of meaningful sentences,” lead author David Moses said. “We’ve shown that it is actually possible to facilitate communication in this way and that it has potential for use in conversational settings.”
Chang said translating the signals intended to control muscles of the vocal system — rather than the signals that move limbs to allow typing — is much more efficient.
“Going straight to words, as we’re doing here, has great advantages because it’s closer to how we normally speak,” Chang said.
Two Harvard neurologists, meanwhile, billed the study as a “pioneering demonstration” in an accompanying editorial in the New England Journal of Medicine, saying the device could potentially help people with amyotrophic lateral sclerosis, cerebral palsy, stroke or other disorders.
“Ultimately, success will be marked by how readily our patients can share their thoughts with all of us,” neurologists Leigh Hochberg and Sydney Cash wrote.
With Post wires
Leave a Reply