A brain-computer interface (BCI) implant from UC Davis has enabled an ALS patient to speak independently for more than 3,800 hours over the past two years, achieving 99 percent and allowing him to work fully. The participant, Casey Harrell, a 47-year-old man with ALS, has produced nearly 2 million words at an average speed of 56 words per minute. This study, published in Nature Medicine, represents the longest-running demonstration of a brain-computer interface (BCI) functioning as a practical everyday communication tool outside of the laboratory.
See also: Neuralink Blindsight: FDA approval for vision restoration implant
The implant consists of four microelectrodes placed in Harrell’s left precentral gyrus, the area of the brain responsible for speech, recording activity from 256 cortical electrodes. Machine learning algorithms in a software platform called BRAND, developed by UC Davis postdoctoral researcher Nicholas Card, translate this neural activity into English phonemes, which are then mapped to words and sentences. The system synthesizes Harrell’s pre-ALS voice to read the decoded text.
In controlled tests with a vocabulary of 125,000 words, the system achieved over 99% word accuracy. In everyday use outside the lab, Harrell rated 92% of sentences as accurate or mostly correct, communicating more than 183,000 sentences during the study period.
Neurosurgeon David Brandman, who implanted the device in 2023 and co-led the study, stressed the importance of being able to communicate on a daily basis for people who want to speak but can't. Despite the paralysis, Harrell has returned to full work and can have meaningful conversations with his daughter, who has never heard his voice.
The significance of the study lies not only in its accuracy but also in the independence it provides. Previous BCI systems required the presence of researchers every time the device was in use or required the patient to be transported to a laboratory. Harrell's system operates from his care team in his home without the need for support from researchers.
See also: Brain implant partially restores movement to quadriplegic man
On average, Harrell used the system for more than five hours a day. The UC Davis team is part of BrainGate, a consortium developing BCIs for speech rehabilitation, computer control, and motion recovery. The hardware used is not custom-made but consists of existing microelectrodes produced by Blackrock Neurotech.
The innovation is attributed to the software, particularly the machine learning algorithms in the BRAND platform that decode speech effort from neural signals in real time.
Brandman compared the current state of BCI technology to the first pacemakers, which required external wiring and large batteries in the 1950s. Today, pacemakers can be implanted in outpatient procedures. He noted that we are still in the early stages of this technology.
Although Harrell is still connected to external computers, advances in AI and hardware miniaturization from companies like Neuralink, Synchron , and Paradromics suggest a future where the device will be less bulky. The competitive landscape in BCI is evolving rapidly, with Neuralink having implanted devices in at least 21 patients under research protocols, although it does not have commercial approval.
See also: The most interesting conspiracy theories after Diddy's arrest
China approved the first commercially available invasive BCI earlier this year.
