PARIS: Advancements in brain implant technology are offering renewed hope for individuals who have lost the ability to speak due to illnesses or accidents, with promising results emerging from two separate studies.
Researchers are making significant strides in addressing communication disorders through the use of brain implants, providing potential solutions for patients who have experienced a loss of speech function.
Pat Bennett, a former dynamic human resources senior executive, was diagnosed over a decade ago with amyotrophic lateral sclerosis (ALS), commonly known as Lou Gehrig’s disease. This neurodegenerative disorder disrupts nerve function, leading to paralysis as the condition progresses. As the disease gradually took away her ability to speak, Pat became a symbol of the challenges faced by those who lose the power of speech due to neurological issues.
Breakthrough in Brain Implants
In a breakthrough study reported in the journal Nature, researchers from Stanford University’s department of neuroscience implanted four small squares, each containing 64 micro-electrodes made of silicone, into Pat’s brain in March of the previous year. These micro-electrodes, located just 1.5 millimeters deep into the cerebral cortex, record electrical signals generated by brain areas associated with language production. These signals are then transmitted outside the skull through cables and processed by an algorithm.
Over a period of four months, the system learned to interpret the signals by associating them with phonemes, the distinct sound units that differentiate words. The signals were processed using a language model, allowing Pat to use a brain-computer interface (BCI) machine to communicate through a screen at a rate exceeding 60 words per minute. Although this rate falls short of a standard conversation, it marks significant progress from previous attempts. Moreover, the error rate for a 50-word vocabulary has dropped to below 10 percent.
A second study, led by Edward Chang and his team at the University of California San Francisco, employed a device with 253 electrodes placed on cortical material. This approach achieved a median rate of 78 words per minute, representing a substantial improvement for the patient, who was able to communicate only at 14 words per minute through head movement interpretation.
Chang’s system uniquely examines signals not only from areas directly linked to language but also from the sensorimotor cortex, which controls facial and oral muscles. This innovation allows for the production of language in various forms, including text, synthetic voice, and an avatar that mirrors a patient’s facial expressions during speech.
The researchers are exploring the development of a wireless version of the implant, which could significantly enhance patients’ independence and social interactions. These advancements hold immense promise for individuals who have lost their ability to speak, offering the potential to restore not only words but also a crucial aspect of identity through communication.
