|The researchers' AI-powered application can read the sentences tetraplegic patients construct /Photo Credit: Peter Porrini (via Shutterstock)|
In the digital age, handwriting is “becoming a rare skill,” but researchers discovered a new tool that would help tetraplegic people communicate, as stated by Daphne Leprince-Ringuet of business technology news site ZDNet. At the Society of Neuroscience’s annual meeting in Chicago, neurologists presented a new application that has the capability of reading the sentences constructed by a volunteer paralyzed from the neck down.
The tool can read the sentences “in double the average speed recorded for existing technologies." The volunteer for the research was instructed to imagine that he was moving his arm to write each letter of the alphabet using an imaginary pencil. As a movement, writing requires “a certain cerebral organization” that occurs in the primary motor cortex. This was used by the researchers to develop their new tool.
An attempt to write, even if it was done mentally, triggers different types of brain activity. These activities can be recorded via microelectrodes and used “to train a neural network.” The network was incorporated into a brain-computer interface (BCI) to recognize the brain’s attempt to write, which would be translated into text in real-time. The computer managed to read the volunteer’s sentences with “92% accuracy at a speed of 66 characters per minute.”
Mistakes occurred when letters looked similar such as “g” and “q.” The researchers acknowledged that the aforementioned letters require more effort. Despite that, it is twice as fast as other technologies. One technology decrypted a tetraplegic’s attempt to communicate at 39 characters per minute.
The researchers concluded, “These preliminary results suggest that a handwriting BCI could be accurate enough to achieve high communication rates.” According to the researchers, the tool needs to refine its translations and translate texts at even greater speeds. However, BCI technology is cumbersome, as it can only be performed within specialized laboratories. BCI also needs an in-house team of engineers and neurophysiologists.
The new handwriting tool is innovative and game-changing, but it would take before it is fully implemented in the health sector.