November 19, 2012
Researchers Develop Better Algorithm For Thought-Controlled Computer Cursors
[ Watch the Video: Controlling A Cursor With Your Mind ]
redOrbit Staff & Wire Reports - Your Universe OnlineA team of researchers from Stanford University have developed a new algorithm that they say can improve both the speed and accuracy of neural prosthetics, allowing paralyzed computer users to have greater control over their thought-controlled cursors.
According to the university, the ReFIT algorithm, as it has been dubbed, features a silicon chip that is implanted into a person's brain and records "action potentials" in neural activity. Those "action potentials" originate from an array of electrode sensors, and once recorded, the data is transmitted to a computer.
"The frequency with which action potentials are generated provides the computer key information about the direction and speed of the user's intended movement," officials from Stanford explained. "The ReFIT algorithm that decodes these signals represents a departure from earlier models. In most neural prosthetics research, scientists have recorded brain activity while the subject moves or imagines moving an arm, analyzing the data after the fact."
In order to determine how the ReFIT system actually worked, they tested it against existing algorithms in side-by-side demonstrations conducted under "closed-loop control conditions." Those demonstrations involved rhesus monkeys, which neurally directed a cursor towards a target on the computer monitor screen, with computers analyzing and implementing visible feedback that was gathered in real time, as it happened.
Writing in the journal Nature Neuroscience, lead author and Stanford University professor of electrical engineering, bioengineering and neurobiology Krishna Shenoy and colleagues reported that the ReFIT algorithm worked twice as well as existing systems.
Shenoy claims that their system was closing in on a performance-level rivaling that of an actual arm, and reported that the system was still performing at a high level more than 48 months after implementation. In comparison, traditional algorithms experienced what the university called "a steady decline in performance over time."
ReFIT "is able to make adjustments on the fly when while guiding the cursor to a target, just as a hand and eye would work in tandem to move a mouse-cursor onto an icon on a computer desktop," the university said. "If the cursor were straying too far to the left, for instance, the user likely adjusts their imagined movements to redirect the cursor to the right. The team designed the system to learn from the user's corrective movements, allowing the cursor to move more precisely than it could in earlier prosthetics."
"Critical to ReFIT's time-to-target improvement was its superior ability to stop the cursor. While the old model's cursor reached the target almost as fast as ReFIT, it often overshot the destination, requiring additional time and multiple passes to hold the target," they added. "The key to this efficiency was in the step-by-step calculation that transforms electrical signals from the brain into movements of the cursor onscreen."
Shenoy, research associate Dr. Vikash Gilja, bioengineering doctoral candidate Paul Nuyujukian, and colleagues developed a method to teach the algorithm about movement. As one of the monkeys moved his actual arm in order to move the cursor, the computer was able to match those movements with its neural activity using computer signals. Later, when the monkey thought about moving the cursor, the computer was able to take that neural activity and translate it into actual, onscreen cursor movement -- a process which was later refined for improved accuracy.
The team is now looking to test the ReFIT device as part of a planned, forthcoming clinical trial. Gilja is optimistic about the algorithm, saying that he believes that the research team as "a good chance" of delivering a product that will be "very useful" to the paralyzed individuals seeking easier ways to control computer cursors.