Two studies published in the New England Journal of Medicine demonstrate significant progress in using brain-computer interfaces to restore speech in paralyzed patients with amyotrophic lateral sclerosis. These advancements show promise for practical, clinically viable applications, according to Edward Chang, MD, a neurosurgeon at the University of California, San Francisco, in an accompanying editorial.
The first study involved a 45-year-old man with severe speech difficulties due to amyotrophic lateral sclerosis (ALS). Researchers implanted four Blackrock Neurotech microelectrode arrays, utilizing 256 intracortical electrodes to record neural activity in brain areas associated with language and speech. This approach used substantially more electrodes than previous studies.
Key features of this neuroprosthesis include:
- Ability to learn rare words
- Rapid training and online recalibration capabilities
- Access to a 125,000-word vocabulary by the second day of use
- Integration with text-to-speech software mimicking the patient's pre-ALS voice
The patient achieved remarkable results:
- Communication rate of 32 words per minute within 16 cumulative hours of use
- Only 2.5% of attempted words incorrectly identified
- Ability to engage in complex conversations, including humor and emotional expressions
Comparatively, smartphone dictation apps have an approximate 5% word error rate, while able-bodied speakers have a 1-2% error rate when reading aloud.
The second study focused on a woman who received a more primitive Medtronic neuroprosthesis at age 58. This device functioned well for six years, enabling communication through clicks. However, progressive brain atrophy due to ALS eventually rendered the interface ineffective, despite no technical malfunction.
Dr. Chang suggests that future efforts may need to target different brain regions less affected by or less prone to degeneration during disease progression.
Several companies, including Blackrock Neurotech, Medtronic, Synchron, and Elon Musk's Neuralink, are working towards commercializing brain-computer interfaces. These studies provide compelling evidence of rapid progress in the field, offering hope for improved communication methods for individuals with ALS and other paralyzing conditions.
