Researchers at the Howard Hughes Medical Institute have successfully turned mental handwriting into on-screen text thanks to a combination of neural implants and Artificial Intelligence.
Working with a 65-year-old man paralyzed from the neck down, the researchers used sensor implants to detect neural activity linked to writing by hand. As the man imagined writing letters, that activity was fed into an algorithm that would translate that activity into text.
“Right now, other investigators can achieve about a 50-word dictionary using machine learning methods when decoding speech,” said Krishna Shenoy, co-author of the study. “By using handwriting to record from hundreds of individual neurons, we can write any letter and thus any word which provides a truly ‘open vocabulary’ that can be used in most any life situation.”
While this is only a proof of concept for one patient so far, it is another promising step towards much-needed independence for those suffering from paralysis and the researchers hope to move on to working with patients who suffer from a degenerative neurological disorder that can take away people’s ability to speak.
Featured image courtesy of HHMI via hhmi.org