The Daily Star Lebanon Breaking News, Lebanon News, Middle East News & World News
A man in the United States with ALS, a muscle degeneration disease, has successfully “spoken” using a brain-computer interface (BCI), bringing new hope to thousands who are unable to speak due to similar conditions.
Casey Harrell, 45, utilized the BCI, which interpreted his brain signals and converted them into text. This text was then spoken aloud by a computer with an impressive 97% accuracy, making this BCI the most accurate speech neuroprosthesis ever reported.
“Our BCI technology enabled a man with paralysis to communicate with his loved ones and caregivers,” said neurosurgeon David Brandman from the University of California, Davis, where the study took place. Brandman described the technology as “the most accurate speech neuroprosthesis ever reported.”
Harrell used the system for both prompted and spontaneous conversations, with real-time speech decoding, according to a release from UC Davis on Wednesday, August 15.
The decoded words were displayed on a screen and read aloud in a voice that closely resembled Harrell’s pre-ALS voice. The voice was generated using software trained on existing audio samples of his voice before ALS affected him.
How the BCI Helped a Man with ALS ‘Speak’
Before the experiment, Harrell’s speech was difficult to understand, often requiring others to interpret for him. In July 2023, sensors from the BCI were implanted in Harrell’s brain. These included four microelectrode arrays placed in the part of the brain responsible for coordinating speech, capable of recording brain activity from 256 cortical electrodes.
The researchers essentially “listened” to the part of Harrell’s brain that was attempting to send commands to his speech muscles. “We’re translating those patterns of brain activity into phonemes—like syllables or units of speech—and then into the words they’re trying to say,” explained neuroscientist Sergey Stavisky. “We’re detecting their attempt to move their muscles and talk.”
The brain signals were then interpreted by machine-learning programs within the device.
Why This Is a Breakthrough
This experiment represents a significant advancement for those who wish to speak but are unable to. “Previous speech BCI systems had frequent word errors, making it difficult for users to be consistently understood, which hindered communication,” Brandman said. “Our goal was to create a system that allows someone to be understood whenever they want to speak.”
“The first time we tried the system, Casey cried with joy as the words he was trying to say correctly appeared on-screen. We all did,” said neuroscientist Sergey Stavisky, who participated in the study.
“At this point, we can decode what Casey is trying to say with about 97% accuracy,” Brandman added.
“It has been incredibly rewarding to see Casey regain his ability to speak with his family and friends through this technology,” said Nicholas Card, the study’s lead author.
Understanding ALS
ALS, also known as Lou Gehrig’s disease, is a progressive disease that causes muscle degeneration. It gradually impairs the ability of nerve cells to control movement, leading to muscle weakness.
Over time, ALS patients lose the ability to stand, walk, and use their hands, and in some cases, it can also affect speech as it impacts the muscles and nerves responsible for speaking.
