Stanford University found a way to make brain-controlled keypads more accurate
A new study from Stanford University found a way to make brain-controlled keypads more accurate.
Thought-controlled prosthesis is becoming a more common solution to help people with spinal-cord injury or disease (SCI/D) and amyotrophic lateral sclerosis (ALS) in their everyday lives. The method bypasses the injury where messages from the brain can’t reach the spine and instead takes the messages directly from the brain and sends them to robotic arms or computer cursors to operate things like a virtual keyboard.
Krishna Shenoy, a Stanford professor of electrical engineering, and her team used this method to test a virtual keyboard that allows people with SCI/D and ALS to use a computer or tablet.
However, because of the complexity of the brain, errors happen and, in turn, slow down the process. Shenoy and her team created an algorithm for measuring the brain signals that makes thought-controlled prosthesis more precise and then tested it on two monkeys.
Researchers showed the monkeys a keypad with blank circles and trained them to reach for the circles that lit up. Using their fingers, the monkeys averaged 29 correct taps every 30 seconds. Shenoy and her team then carried out the same procedure, but instead measured the taps from the monkeys’ brain-controlled cursors and adjusted the algorithm accordingly. With the brain prosthetic, the monkeys scored 26 taps every 30 seconds, about 90% as quickly as using their fingers.
“Brain-controlled prostheses will lead to a substantial improvement in quality of life,” Shenoy says. “ … This is a fundamentally new approach that can be further refined and optimized to give brain-controlled prostheses greater performance and therefore greater clinical viability.”
The U.S. Food and Drug Administration has approved Shenoy’s team to move ahead with a clinical trial to test the same process in people with SCI/D.
(Register or login to add comments.)