By: Allison Kubo
Study participant T5 was paralyzed from the neck down but was transferred to the screen as he imagined writing. Brain-computer interfaces (BCI) aim to restore the function of people who have difficulty or even lose the ability to move or speak. And yes, it would allow you to control a computer with your brain. New research published in Nature on May 12 shows advances in restoring writing for those suffering from paralysis. Study participant T5, in view of this denominator, used the BCI type with high accuracy close to the speed of the control group in order to maintain privacy. They achieved a raw accuracy of 94.1% and an accuracy of more than 99% with an auto-correction function.
How does this work? Studies have shown that thinking about the movement you want activates the necessary parts of the brain even after years of paralysis. The researchers instructed T5 to visualize writing with a pen on lined paper. The brain gave its signals, but something in communication, such as a spinal cord injury, prevented it from moving towards the hand. However, by capturing the brain signals using two microelectrodes positioned on the skull near the brain that control the movement of the right hand (precentral gyrus), the signal can be translated into words by the computer rather than the hand. Of course, it would be impossible to decode just a raw electromagnetic signal from the brain. By training the model with multiple recordings of known signals, the researchers can match the signals with known letters.
Why does it work This recent study increased typing speed to 90 words per minute. Compared to point-and-click input or eye movement methods measured at only 40 characters per minute, this offers a significant improvement. With point-and-click methods, people move a cursor on a screen and recognize the signals in a similar way. Although this method may be more diverse in that it can be applied to games, websites, etc., it is limited by how fast the computer can decode the signals it senses. This is far slower than decoding handwriting signals. One possible explanation is that writing is a far more diverse signal than moving a cursor in a straight line. The curves of the letters vary over time, but moving a cursor is a more constant movement in a single direction.
The method is not yet generally available to those who need it. However, it offers a new option to restore valuable communication for those suffering from spinal cord injuries or diseases like ALS.