A 47-year-old man with ALS is able to communicate independently again thanks to a brain implant. It is a Brain-Computer Interface (BCI), through which he has already spoken nearly 2 million words.
This is according to a study published in *Nature Medicine*. Casey Harrell, a 47-year-old man with ALS, is a participant in the BrainGate2 clinical trial. He suffers from tetraparesis (weakness in the arms and legs) and dysarthria (slurred speech). In 2023, neurosurgeon David Brandman of UC Davis implanted an experimental BCI device in Harrell’s left precentral gyrus, a brain region involved in coordinating speech. The system consists of four micro-electrode arrays that record the activity of 256 cortical electrodes.
Nearly 2 million words
For nearly two years, Harrell has used the BCI system almost daily at home. During that time, he communicated more than 183,000 sentences and nearly 2 million words, and used the system for more than 3,800 hours. His average communication speed was 56 words per minute.
The BCI system enables Harrell to send emails and messages, surf the internet and communicate with family and friends. “It’s a life that’s more filled with dynamic action, with friends and family, with colleagues, and it’s something that allows me to communicate in a way that’s closer to my natural way of communicating than any other technology I’ve ever experienced,” says Harrell.
Translating neural signals into text
The BCI technology was developed by scientists at the University of California, Davis (UC Davis), in collaboration with colleagues from Brown University and the Mass General Brigham Neuroscience Institute. The technology uses advanced decoding algorithms. These translate neural signals into text (speech-BCI) and enable cursor control (movement-BCI), allowing users to operate a computer independently. Thanks to this combination, Harrell is able to continue working.
“Casey can use the system to communicate his own thoughts, not just when we are there in a controlled environment, but whenever he wants to. Sometimes he does so for up to 12 hours at a time,” explains Nicholas Card, a postdoctoral researcher at the UC Davis Neuroprosthetics Lab.
High accuracy
The study shows that Harrell rated 92 per cent of the sentences as accurate or largely correct. In controlled tests, the system achieved a word accuracy of over 99 per cent with a vocabulary of 125,000 words.
“In our previous study, we demonstrated 97% accurate word decoding. But Harrell could only use the neuroprosthesis if someone from our research team was present to set it up. We have now made improvements that bring this medical technology closer to clinical use: he can use it at home without researchers. It is even more accurate (99%), keeps up with him when he tries to speak faster, and has been working excellently for almost two years,” said Sergey Stavisky, neuroscientist and co-director of the UC Davis Neuroprosthetics Lab.
The researchers regard BCI technology as a significant step forward. Brandman: “For years, BCIs were proof-of-concept devices that only functioned in highly controlled research environments. This work demonstrates that we may have crossed a threshold by enabling someone with paralysis to speak on their own terms.”
Image: Regents of the University of California, Davis
