News release

From: Springer Nature

Brain implant restores communication independence at home

An intracortical brain–computer interface (BCI) that enables long-term, independent speech-based communication and computer use in an individual with severe paralysis is reported in Nature Medicine. The findings address longstanding challenges that have limited high-performance BCI usage to highly controlled settings, and show that they could be used as practical assistive tools in everyday life.

Loss of speech and the ability to control devices is a common result of conditions such as amyotrophic lateral sclerosis (ALS), and it often reduces independence and quality of life. Existing assistive technologies can help but are frequently slow, unreliable, or require extensive support from trained professionals. While previous studies have shown high accuracy for speech or cursor control in controlled settings, few have enabled independent use at home, and none have supported speech, or both speech and computer control together.

Nicholas Card and colleagues describe the near-daily home use of a multimodal BCI over almost 2 years by a man with severe paralysis and speech difficulties caused by ALS. Neural signals recorded from implanted electrode arrays in the speech motor cortex of the brain were decoded in real time into text and cursor control. The participant used the system for more than 3,800 hours, communicating 183,060 sentences (almost 2 million words) at an average of 56 words per minute, with 92% of the sentences rated at least mostly correct. In structured testing, word accuracy exceeded 99% from a vocabulary of more than 125,000 words. Sustained performance was enabled by faster-calibrating decoders, adaptive software features (including background calibration and gaze-based controls), and a streamlined startup process that allowed care partners to operate the system independently.

The findings demonstrate that intracortical BCIs can provide stable, high-performance communication without researcher assistance over extended periods. However, the study involved only a single participant, and the system still requires some training on use by carers. Further work is needed to assess wider applicability, improve portability, and achieve consistently high accuracy during natural conversation.