Stable Decoding from a Speech BCI Enables Control of Assistive Devices Without Model Retraining in ALS
Summary
This study demonstrates that a chronically implanted ECoG-based speech BCI can reliably control assistive devices over a 3-month period without model retraining or recalibration. A participant with severe dysarthria due to ALS operated computer applications with six intuitive speech commands via a chronic electrocorticographic implant over the ventral sensorimotor cortex, achieving median accuracy of 90.59%.
Why it matters
The plug-and-play capability of this BCI system eliminates the need for daily recalibration, making it practical for independent home use by individuals with ALS. The long-term stability of decoding without model retraining represents a critical step toward translating research-grade BCIs into clinically viable assistive communication technologies.
Want the full NeurotechMag Pro intelligence database?
Access 800+ papers, trials, patents, funding rounds and news items — updated daily.
Start free trial