Software that translates eye movements into commands will allow people with mobility disabilities to control various devices / Rachel Nauer
The article was published with the approval of Scientific American Israel and the Ort Israel network
In early 2015, when Eric Sorto, paralyzed from the neck down, managed to control a robotic arm with the power of thought to direct it to put a can of beer to his lips, the media went wild. There is no doubt that it was an impressive achievement. However, the technology behind this achievement, a chip loaded with electrodes implanted in Sorto's brain, is an expensive, invasive technology that often requires months of training. And worse: this technology requires a mental and physical profile that only a few paralyzed people have.
There may be a better way than this. Instead of making a direct connection between the electrical activity in the brain and the device you want to operate, Aldo Faisal, professor of neurotechnology at Imperial College London, suggests using eye movements to control wheelchairs, computers and computer games. Using simple commercial video game cameras, Faisal and his partners built glasses that track the user's eye movements and feed the data to the computer. Software translates this data into commands that control the device. Almost anyone can use this technology: amputees, Parkinson's patients, multiple sclerosis patients and people suffering from muscular dystrophy. Building the system costs less than $50. Of the thousands who volunteered to try the system when it was presented at a science fair, the vast majority mastered its operating principles within 15 seconds at a level high enough to play the famous video game Pong, and this without receiving any instruction.
Scientists have known for years that eye movements can reveal the immediate intentions of humans: where they intend to walk, what they intend to do, whom they intend to approach. Faisal and his colleagues used the knowledge gained in 70 years of eye movement research to develop algorithms that turn a change of gaze into a command to control a wheelchair, a wink to press a mouse key, or a pupil movement to swing a game racket. To predict the user's intent, the researchers train the algorithms using data previously obtained from recording the eye movements of volunteers while driving a wheelchair using a steering rod or robotic arm. The software gradually learned to recognize the difference, say, between the eye movements people make when they look at a mug to check what's inside and their eye movements when they intend to pick it up to sip from it.
Before Faisal can integrate his system into commercial medical products, he must raise funding for clinical trials. Until that happens, a four million euro grant from the European Union will allow his research group to develop wearable exoskeletons that paralyzed people can control using the eye movement system the group developed. "I want to see what I can do to help people walk again," Faisal says. "That's what I focus on."
One response
The beginning of this was in the 60s of the last century as part of the Apollo program. During the launch of each Apollo spacecraft, for a few seconds they had to be under a load of 12g. The astronauts were trained for each flight what they should do at each moment. Knowing that they would encounter such a high load, they learned how to operate various devices when needed with the help of their eyes.