Direct Brain to Text

Imagine after years of not being able to communicate, seeing letters you picture in your mind appearing on a computer screen in real time. This is the latest attempt at connecting disabled people with the world through handwriting.

But how when they can neither use their hands nor speak? Because the neural processes for those skills are not lost.

A team at Stanford University first relied on implanted brain sensors to follow a patient's thoughts about arm movements to move a cursor on a screen to point and click on letters. It was slow at about forty characters a minute. Then they wondered whether the neural activity associated with handwriting would work.

So, they chose a 65-year-old man, paralyzed from the neck down, who had a brain computer interface or BCI implanted in the region of the brain that controls the hand and arm. As the man imagined writing letters, the BCI sensors received those signals from individual neurons, and a machine learning algorithm recorded them. He was able to copy sentences and answer questions at about ninety characters per minute which is almost as fast as someone his age typing on a smart phone.

This "Brain-To-Text" BCI proved to be so fast because the neuronal activity for each letter was highly distinctive making it easy for the algorithm to identify each one.

For people in desperate need of connection, this innovation is monumental. Let's hope they don't have to wait too long