A team of scientists from the University of California has reported a breakthrough in the field of neurocomputer interfaces (BCI). They have managed to create a device that can translate brain signals into text with unprecedented accuracy—an error rate of less than 3%.

Casey Harrell, a 45-year-old American who has amyotrophic lateral sclerosis (ALS), also known as Lou Gehrig’s disease (Lou Gehrig, ALS), participated in a study published in the New England Journal of Medicine. The disease, which began five years ago, deprived Harrell of full communication. If the average speech rate of healthy people is around 160 words per minute, Harrell could speak an average of 6 words per minute. His speech was very slow due to the loss of motor neuron function specific to this disease.

But thanks to new technology, Harrell was able to regain the ability to communicate. During the procedure, he was implanted with 3.2 mm microelectrode arrays, a signal processing system based on BlackRock Neurotech’s NeuroPort technology. The system transmits brain signals to several computers with special software Backend for Realtime Asynchronous Neural Decoding (BRAND), which decodes neural signals in real time and then displays them on the monitor screen as phrases and sentences.

Already in the first session, when Harrell tried to pronounce a sentence from a 50-word dictionary, the decoding accuracy was 99.6%. In the second session with the same dictionary, all sentences were decoded without errors. The dictionary was later expanded to more than 125,000 words, covering most of spoken English. After a few hours of training, the decoding accuracy reached 902%, and in the following months it exceeded 975%.

The study was presented by a group of scientists led by neuroscientist Sergey Stavisky and neurosurgeon David Brandman at the University of California, Davis. Although Harrell was the first to test the new neuroprosthesis and interface technology, the results show great promise for restoring communication skills in people with disabilities.https://www.youtube.com/embed/thPhBDVSxz0?si=qyEIaj4kHT3qOTfX

When When we first tested the system, he (Harrell) cried tears of joy as the words he was trying to say appeared on the screen very quickly. But we were all very moved. “, – noted Stavinsky.