Brain Waves of a Paralyzed Man Converted to Speech in a World-First Breakthrough
According to a scientific report published Thursday, US researchers have built a neuroprosthetic device that successfully translated the brain signals of a disabled man into whole phrases.
“This is an important technological milestone for a person who cannot communicate naturally,” said David Moses, a postdoctoral engineer at the University of California, San Francisco (UCSF) and one of the study’s lead authors.
“It demonstrates the potential for this approach to give a voice to people with severe paralysis and speech loss.”
The discovery included a 36-year-old man who suffered a stroke at the age of 20 and was left with anarthria – the inability to speak coherently, despite the fact that his cognitive function remained intact.
Thousands of people lose their capacity to communicate each year as a result of strokes, accidents, or disease.
Previously conducted research in this field concentrated on reading brain waves via electrodes in order to construct mobility prosthesis that enable users to type out letters.
The new technique was designed to facilitate more spontaneous and fast dialogue.
UCSF researchers previously implanted electrode arrays on individuals with normal speech who were having brain surgery in order to decipher the signals that regulate the vocal tract’s expression of vowels and consonants and evaluate the patterns to anticipate words.
However, the notion has not been tested on a paraplegic patient to demonstrate clinical benefit.
Achievement in neuroengineering
The researchers decided to start a new study called Brain-Computer Interface Restoration of Arm and Voice, and the first participant requested to be called BRAVO1.
BRAVO1 has severely limited head, neck, and limb motions as a result of a devastating brainstem stroke and communicates by poking letters on a screen with a pointer affixed to a baseball cap.
The researchers worked with BRAVO1 to establish a 50-word vocabulary that included everyday terms such as “water,” “family,” and “good,” and then implanted a high-density electrode over his speech motor brain via surgery.
Over the next several months, the scientists recorded his cerebral activity as he attempted to pronounce the 50 words and utilized artificial intelligence to detect and correlate tiny trends in the data.
They provided him with sentences generated from the vocabulary set and recorded the outcomes on a screen to verify that it worked.
They then prodded him with questions such as “How are you today?” and “Would you like some water?” to which he responded with “I am very good,” and “No, I am not thirsty.”
With a median accuracy of 75%, the system decoded up to 18 words every minute. Its success was aided by a “auto-correct” feature akin to that found on smartphones.
“To our knowledge, this is the first successful demonstration of direct decoding of full words from the brain activity of someone who is paralyzed and cannot speak,” stated co-author and BRAVO1 neurosurgeon Edward Chang.
The journal’s accompanying editorial lauded the breakthrough as a “a feat of neuroengineering,” and indicated that technological developments such as smaller surface electrodes could assist enhance accuracy even more.