Researchers recently reported, with voice silenced for years, an extensively paralyzed man was able to communicate, using technology that decodes electrical impulses produced by his brain each time he attempts to speak.
According to a report from The Washington Post, the advance that the University of California at San Francisco announced, is believed to mark the first time an individual has restored the power to communicate using words and short sentences to someone who had lost it due to neurological damage.
The 38-year-old man, who opted to remain unnamed but is dubbed BRAVO-1 in the research, suffered a brain stem stroke 15 years ago that severed the neural link between his brain and his vocal cords.
As indicated in this report, he is paralyzed from the neck down and has been communicating by thoroughly tapping letters on a keyboard with a pointer attached to a baseball cap's bill.
A Computer that can Decode More Than 15 Words Per Minute
Now, simply by trying to utter words, the man has 50 at his disposal and can develop short sentences that primarily concern his welfare and care.
A computer decodes the activity of his brain and shows the sentences on a screen with median preciseness of roughly 75 percent, at a rate of more than 15 words a minute. The average conversational speech takes place at roughly 150 words a minute.
Neural engineering assistant professor Christian Herff, at Maastricht University in the Netherlands, who was not part of the new study, called this progress described in the research as "gigantic".
Past research had shown the same technique in the test subjects who still were able to speak. Describing the development, Herff said it is actually quite a big deal. This is the first study that certainly does it in someone who is unable to speak.
'Neuroprosthesis Technology'
According to the research team's leader, Edward Chang, chairman of the Department of Neurological Surgery of UCSF, the advance would not have been possible even five years back.
Since then, such progress in artificial intelligence, as well as the decoding of neural signals led to the result of the study, "Neuroprosthesis for Decoding Speech in a Paralyzed Person with Anarthria", published in The New England Journal of Medicine. The authors described such a technology as a "neuroprosthesis".
In an interview, Chang said that he has been working in this area for a decade, motivated by the patients he observed, who had lost the ability to speak.
Thousands of people experience that fate every year as a result of trauma, strokes, and diseases like amyotrophic lateral sclerosis or ALS, and cerebral palsy.
Brain Signals Generated
Chang entrenched a grid of electrodes on the patient's brain's sensorimotor cortex, which controls the speech's production.
A wire is carrying the electrical signal from the electrodes to a port that's attached permanently to the top of his head and can be connected to a computer by a cable.
During 48 sessions that lasted 22 hours, the scientists recorded the brain signals BRAVO-1 generated as he tried to say 50 words displayed on a screen.
In their paper, the researchers wrote, they then used deep-learning algorithms to develop computational models for the detection and classification of words from a pattern in the cortical activity earlier recorded.
They used other models to produce the probable next words in sentences the man was attempting to say. Chang said mere chance would have led to a two-percent preciseness rate with a 50-word vocabulary.
As a result, the study authors were able to generate correct words and sentences as much as 93 percent of the time.
Most of the study in this portion of the field of brain-computer interface has been carried out on patients suffering from epilepsy who volunteered after having electrodes implanted to diagnose their seizure's source.
Chang and other researchers believed a person who has anarthria, described as Healthline as the incapability to speak, still would be able to produce the same activity of the brain, although it was not sure until his team succeeded.
Related information about brain signals is shown on JAMA Network's YouTube video below:
RELATED TOPIC : Your Happiness Might Very Well Be Inherited, Says Scientists
Check out more news and information on Brain in Science Times.
