July 21

Scientists connected a dumb man’s brain to a computer, and now he can speak


In addition, they can also develop wireless communication technologies between electrodes in the brain and computers. If successful, this could eliminate BRAVO-1’s dependence on the cord connecting the back of his skull.

For the first time in medical history, doctors were able to help a man who had been paralyzed and mute for many years talk again by translating his brain waves into speech.

The 38-year-old man, living in the US, was nicknamed BRAVO-1 in research conducted by the University of California. Fifteen years ago, this man suffered a stroke that severed the nerve connecting his brain to his larynx.

BRAVO-1 is also paralyzed from the neck down, so it can only use head movements to communicate. A team of engineers designed him a special chair with a touch screen connected to a computer in front of him.

Normally, BRAVO-1 would wear a cap with a long stick that would allow him to touch the tip of the sensor to the screen and choose words to say.

But now, with an electrode attached directly to the inside of the brain, BRAVO-1 can speak without the need for a helmet. His nerve impulses were transmitted directly to the computer for decoding, allowing him to speak at an average speed of 15 words per minute with 74% accuracy.

At times, the speed can be pushed to a maximum of 18 words/minute, and the accuracy reaches 93%. In comparison, the average person would speak at 150 words per minute and achieve 100% accuracy.

That shows that BRAVO-1’s ability to decode brain waves into speech is still not really impressive. But the scientists say that the first word BRAVO-1 uttered itself was “an important technological milestone for someone who cannot communicate naturally” like him.

“It shows the potential of this approach, which will help give voice to people who are severely paralyzed and unable to communicate,” said study author Dr. David Moss, an engineer at the University of California, Los Angeles. told the Washington Post.

Helping paralyzed people regain the ability to communicate

To be able to help BRAVO-1 speak again, scientists at the University of California implanted a grid of electrodes on his sensory cortex, the same area of ​​the brain that controls his speech. The electrode is then connected outside the skull via wires into a decoding computer.

Next work was to train a program that could recognize and distinguish electrical impulses from BRAVO-1’s sensory cortex, helping him translate thoughts into speech. This was done through 48 testing sessions lasting a total of 22 hours.

In it, the scientists attempted to record the brain signals BRAVO-1 produced when he spoke 50 words that flashed across the screen. They then used a deep learning algorithm to try to filter out which signals represent which words BRAVO-1 meant.

To speed up his communication, part of the algorithm will allow the computer to guess the next word that BRAVO-1 might come up with. Basically, this algorithm works like a suggestion or spelling correction function on your phone.

However, the scientists say that the normal error correction function gives only about 2% accuracy. In this study, they were able to push that accuracy to 93% with 50 words. It is a small language store but contains all the vocabulary necessary for the daily life of a paralyzed person like BRAVO-1, such as “water”, “family”, and “good”.

During a demonstration to test whether the system was working, the researchers asked him questions like “How are you today?” and “Do you want some water?”.

As a result, BRAVO-1 took a few seconds to think and say on the screen responses like “I’m fine” and “No, I’m not thirsty”.

“To our knowledge, this is the first demonstration that directly decoding the brain activity of a paralyzed person who cannot speak into full words can be done successfully,” said anatomist. Neurologist Edward Chang, co-author of the study said.

Electrodes implanted in the cerebral cortex allow paralyzed people to communicate by thinking with computers

Five years ago, no one thought such a study would be successful

Chang is currently chair of the Department of Neurosurgery at the University of California. He emphasized that five years ago, no one would have thought that a study like this would have such good results.

Most similar studies then, with a computer hooked up to a person’s brain, were done only on patients with epilepsy. This brain-computer interface is only used to diagnose the origin of their seizures.

Going back five years, though, some researchers were able to decode sounds or syllables from a patient’s brain waves. But the accuracy of the algorithms was then too low to be able to combine them into complete words.

The success Chan and his team have achieved with BRAVO-1 today inherit a lot from the field of artificial intelligence over the past 10 years. For example, speech recognition and spelling correction algorithms on computers and mobile phones have also been developed with billions of hours of training on billions of data samples.

In the field of decoding neural signals, scientists have also made a lot of progress. In May, for example, a team at Stanford University helped a paralyzed man write words on a screen using brainwave decoding technology.

In this case, the man imagined how he moved his hand, controlling his fingers to write, instead of thinking the letters would be typed out or pointed at the touch screen like BRAVO- first.

Neurosurgeon Edward Chang (right), author of the new study

Commenting on the new study, Christian Herff, an associate professor of neuroengineering at Maastricht University in the Netherlands, said what Chang and his colleagues did was a “huge” step forward. “It really solved a big problem,” Herff said.

Previous studies have shown that decoding brain waves can help us read the mind of a person who can speak. “But this is the first study to do that in a non-verbal patient,” Herff emphasized.

In an interview, Chang said his new work is the culmination of a decade-long effort. Ten years ago, he met a patient who was mute and began to wonder about the inconvenience this patient had in life.

Thousands of people have suffered the same fate each year due to neurological damage from stroke, cerebral palsy, trauma and diseases like multiple sclerosis – ALS like scientist Stephen Hawking.

“Every day, I face these patients, who have lost the ability to speak after a stroke or brain injury. Silence has wreaked havoc on their lives. After all, words are words. It’s part of what makes us who we are. When you lose it, it’s really devastating,” Chang said.

But with the success of BRAVO-1 today, he will help his patients regain the ability to communicate. “And really this is just the beginning,” Chang said.

The next step, Chang and colleagues will need to improve the deep learning algorithms to be able to increase the accuracy and speed of speaking for BRAVO-1. He expected he would also need to expand his patient’s vocabulary so that he could cover more topics.

Meanwhile, Herff suggested the team could route BRAVO-1’s language into audio. Which means it will allow him to have a real voice, like the old Stephen Hawking but with added intonation and expression.

In addition, they can also develop wireless communication technologies between electrodes in the brain and computers. If successful, this could eliminate BRAVO-1’s dependence on the cord connecting the back of his skull.

The application of a wireless brain-computer interface will also be much broader. It can be suitable even for people who are mute but not paralyzed, who do not need to sit at their computer to speak.

“Although the study counts as a success, we can’t say we’ve accomplished anything with this technology. It’s really just the beginning,” Chang said.

The study was published in the New England Journal of Medicine.


Brainwave, Community Network, research, technology

You may also like

What if we spread solar cells to fill the Sahara?

Get in touch

0 of 350