Human beings are fascinated by the thoughts of someone or something that can read our minds, be it a telepath, computer, or Santa Claus. Now, scientists say they have developed a system that combines machine learning and a brain-computer interface, or BCI, to read handwriting that occurs in the brain rather than on paper.
A team of scientists working with a 65-year-old paralyzed from the neck down using sensors implanted in the brain to detect neurological activity linked to hand writing. As the volunteer imagines writing a letter, the activity is sent to an algorithm that translates in real time into the message displayed on the screen.
Details of the trial are in a report in the current journal Nature, study co-author Krishna Shenoy, a Howard Hughes Institute of Medicine researcher at Stanford University, said the method appears to be more effective than similar studies that tried. doMore than handwriting
“Now other researchers are able to achieve a 50-word dictionary using machine learning methods when decoding speech,” Shenoy said in a statement. “By using handwriting to record hundreds of neurons, we can write any letter and hence any word given. A truly ‘open vocabulary’ that can be applied to any situation in life.
The hope is that such a system can allow people with paralysis to type and. It has a reputation for working on similar brain transplant technology.
“Think how much your day is spent on computers or communicating with others,” Shenoy said. “Restoring the ability of those who have lost their freedom to interact with computers and others is absolutely critical. And that’s what brings a project like this in the middle. ”
The system has proven to be a proof of concept in a single patient until now. Builds on a previous work done by Shenoy’s team, where participants with nerve sensors implant try to move their arms to move the on-screen cursor to point and click on letters and spell words and sentences. This method allows people to “type” about 40 characters per minute, while the new handwriting method is estimated at 90 characters per minute, believed to be a new record for printing via BCI.
The researchers note that the rates were similar to those of the 65-year-old participants who typed on their smartphones.
“Communication is at the heart of our work in society,” said Debara Tucci, director of the National Institute on Deafness and Other Communication Disorders, one of the study’s funders. “In today’s world of Internet communication, people with severe speech and physical disabilities can face significant communication barriers and possibly loneliness. Commercialization of this latest BCI technology “
Shenoy wanted to see handwriting incorporated into more complex BCI-based systems, which also allows speech decoding and point-and-click navigation, as well as the smartphone interface in the present
“Having two or three modes and switching between them is something we naturally do,” he said.
Next, the team aims to work with participants who are unable to speak or suffer from degenerative neurological disorders, such as amyotrophic lateral sclerosis that can rob a patient with a pooled patient. Can