In a groundbreaking discovery, researchers from UC Davis Health have developed a brain-computer interface that can transcribe brain signals to speech for a patient at a 97% accuracy rate. That is a rescue home to the thousands who have lost their voice to amyotrophic lateral sclerosis (ALS) or other paralyzing disorders.
The new system was tried out on Casey Harrell, 45, of Cadiz, Ky., who was diagnosed with the illness ALS and could do little but blink because of it. Days after the BCI was turned on, Harrell was able to think his thoughts. It was a breakthrough in neuroprosthetics. Their findings were published Wednesday in the New England Journal of Medicine.
ALS, known as Lou Gehrig disease, is a progressive neurodegenerative disorder that kills nerve cells controlling muscles, eventually leading to what is described as severe physical limitations and finally speech loss. Brain signals that intend to form speech will be now translated into text by the new BCI and read by the computer.
“The BCI technology from our center helped a paralyzed man to communicate with friends, families, and caregivers,” said David Brandman, a neurosurgeon at UC Davis and co-principal investigator of the study. “Our paper demonstrates the most accurate speech neuroprosthesis ever reported.”
The system operates through a device that identifies the brain activity a user makes when trying to speak and into text, scrolling on a computer. From there, the computer speaks the text. Harrell, with extreme speech impairment, willingly participated in the BrainGate clinical trial. Four microelectrode arrays were embedded into his brain. They measured the amount of brain activity through 256 cortical electrodes in the left precentral gyrus, an area connected to the coordination of speaking.
“We’re picking up their attempt to move their muscles and talk,” said Sergey Stavisky, a co-director of the UC Davis Neuroprosthetics Lab. “We’re recording from the part of the brain that’s trying to send those commands to the muscles and translating those patterns of brain activity into phonemes and words.”
Although BCI technologies had already received progressive development systems had become notoriously slow and inaccurate earlier. New System The new system boasted efficiency and accuracy which systems have never achieved before. In first training session started with 99.6 percent word accuracy on a 50-word vocabulary within 30 minutes. In later sessions” said words increased to 125,000 with an accuracy rate of 97.5 percent.
“The first time we tried the system, he cried with joy as the words he was trying to say correctly appeared on-screen. We all did,” Stavisky said.
Harrell spent 248 hours with the BCI over 32 weeks, during which he was extensively tested with the BCI in various conversational environments, either in person or using video chat. The system’s accuracy in decoding his intended speech has been far better than, to breathe new life into those who struggle to communicate.
“Not being able to communicate is horribly frustrating and incredibly demoralizing. You’re trapped,” Harrell said. “Something like this technology will put people back into life and society.”
“Lead author Nicholas Card said, “It was so rewarding reacquiring the ability for Casey to speak with loved ones. Casey and our other BrainGate participants are truly extraordinary and deserve tremendous credit for joining these early clinical trials,” said co-author of the research, neurologist and neuroscientist Leigh Hochberg.
BrainGate2 clinical trial enrolls participants interested in advancing this life-changing technology. Details can be found at the following: braingate.org, or email: braingate@ucdavis.edu.
This truly monumental breakthrough will represent an even greater stride for neuroprosthetics, thereby giving renewed hope to those who seek to regain their voices.