A brain-computer interface (BCI) technology that converts brain activity into text and reads it out loud has been announced. It is predicted to assist patients with amyotrophic lateral sclerosis (ALS), who’ve difficulty speaking as a result of muscle weakness throughout their body, regain their ability to speak.
Reuters and Bloomberg reported on the 14th (local time) that BCI startup BlackRock Neurotech successfully restored the power to speak by implanting a BCI chip developed by the corporate into the brain of an ALS patient who had difficulty speaking.
ALS, also referred to as Lou Gehrig’s disease, causes patients to step by step lose the power to perform movements as a result of a lack of motor neurons. The muscles used for speech are also affected, making conversation difficult.
Although BCI, which connects the brain and computer, is usually used to revive communication skills in ALS patients, smooth communication has been limited. It is because machine learning programs that interpret brain signals require a considerable amount of data and data processing takes time, making it easy for errors to occur.
The technology detects when a user is attempting to speak by moving their muscles, records the a part of the brain that’s sending the commands to the muscles, and translates that into patterns of brain activity or components of sound to interpret what the user is attempting to say.
Researchers from UC Davis, Massachusetts General Hospital, Brown University School of Medicine, and Blackrock Neurotech conducted a BCI clinical trial on a 45-year-old male patient with ALS, Casey Harrell. The patient, who suffered from quadriplegia and severe speech impediment, was capable of converse together with his caregiver at a median of about 7 words per minute after surgery. A typical conversational rate is about 160 words per minute.
The researchers surgically implanted 4 microelectrode arrays, designed to record brain activity via 256 cortical electrodes, into the world of the brain chargeable for controlling speech.
Based on the researchers, after a complete of 16 hours of use, the BCI chip was capable of speak at a rate of 32 words per minute, and by the second day, its vocabulary had grown to 125,000 words, achieving 90.2% accuracy.
As data gathered, the ultimate accuracy increased to 97.5%, with only 2.5% of the words attempted being misidentified – the very best record ever achieved for the sort of BCI.
When a patient speaks through a BCI created in this fashion, the words decoded by the system are displayed on the screen and browse aloud. Because the patient’s voice before the onset of ALS is used for the voice, patients using the BCI can speak with their very own voice as before.
Meanwhile, Bloomberg reported that when Harrell spoke again, the relations watching were in tears. The primary words he spoke together with his voice back were a repetition of his wedding vows.
Harel also sobbed, saying, “That is real life for me.”
This study Latest England Journal of Medicinewas published in .
Reporter Park Chan cpark@aitimes.com
