Paralysed People Can Now Communicate Twice as Fast Through AI

Artificial Intelligence (AI) has transformed healthcare and rehabilitation, affording people a second chance at independence in ways never before dreamed possible.

One area in which the tech has made great strides is paralysis—particularly for people living with the effects of a stroke or neurological disease, who can have trouble communicating a single sentence.

Famously, in the late ‘90s, Jean-Dominique Bauby—former editor-in-chief of French Elle—was able to write his beautiful and haunting memoir “The Diving Bell and the Butterfly” with just one eyelid after suffering a massive stroke. Claude Mendibil, a specialised nurse, would sit by his bedside for four hours a day, reading out letters of the alphabet and stopping when Bauby blinked. Together, they completed the book in 10 months.

Advances in computer technology have sped up this process. A more recent technique has been to implant electrodes on parts of the brain that control movement, encouraging patients to use their thoughts to move a cursor and select letters on a screen. Such tests reportedly allow users to type 39 characters a minute—faster than blinking, but still about twice as slow as natural handwriting.

 

AI method

And now, using AI, the process is getting even faster. Science Mag reports on a revolutionary experiment in which a volunteer paralysed from the neck down was hooked up to a computer and instructed to imagine moving his arm to write every letter in the alphabet. This brain activity, tracing the trajectory of an imagined pen, trained the computer programme to recognise a series of letters.

The volunteer was then able to imagine writing sentences, which the computer could reproduce with around 95 percent accuracy. The method reached a speed of 66 characters per minute—incomprehensibly faster than blinking; roughly twice the speed of the electrode method; and now, according to Wikipedia, virtually just as fast as regular handwriting (68 characters per minute).

Science Mag adds that this may even get faster, as the technology is refined and researchers further study neural processes to better understand how the brain “plans and orchestrates fine motor movements.”

Leave a Reply

Your email address will not be published. Required fields are marked *