MIT develops brain-to-machine algorithm
Scientists are making progress on neural devices that can translate the thoughts of a paralyzed person into driving action for a prosthetic device
Scientists are making progress on neural devices that can translate the thoughts of a paralyzed person into driving action for a prosthetic device.
Researchers at the Massachusetts Institute of Technology said Wednesday that they've developed an algorithm for a neural prosthetic aid that can link an individual's brain activity to the person's intentions; and then translate that intention into movement.
Of course, other scientists have already done that, and built prototypes for neural brain-to-machine devices that can work for animals or humans. But each team has taken a different approach to the problem, such as developing algorithms for measuring activity in a specific brain region, or measuring them through EEGs vs. optical imaging.
MIT said that it has developed a unified algorithm that can work within the parameters of these different approaches. Lakshminarayan "Ram" Srinivasan, lead author of a paper on the subject, said MIT's new graphical models are applicable no matter what measurement technique is used.
"We don't need to reinvent a new paradigm for each modality or brain region," he said in a statement.
Still, he said, the algorithm isn't perfect, nor the final solution to solving what is a difficult problem. "Translating an algorithm into a fully functioning clinical device will require a great deal of work, but also represents an intriguing road of scientific and engineering development for the years to come," according to MIT.
MIT will publish a paper on the subject in the October edition of the Journal of Neurophysiology.