Major Milestone: Highest-Accuracy, Brain-Controlled Typing for People with Paralysis
Stanford University researchers have demonstrated that a brain-to-computer hookup can enable people with paralysis to type via direct brain control at the highest speeds and accuracy levels reported to date. Their study involved three participants with severe limb weakness; each had one or two small electrode arrays placed in their brains to record signals from the motor cortex, a region controlling muscle movement. These signals were transmitted to a computer via a cable and translated by algorithms into point-and-click commands guiding a cursor to characters on an onscreen keyboard. Each participant, after minimal training, mastered the technique sufficiently to outperform the results of any previous test of brain-computer interfaces for enhancing communication by people with similarly impaired movement. The participants achieved these typing rates without the use of automatic word-completion assistance common in electronic keyboarding applications, which likely would have boosted their performance.
Transcript
00:00:00 [MUSIC PLAYING] Stanford University. The goal of our research collaboration is to restore function to people with paralysis. This can be people who have a spinal cord injury, people who have a neurodegenerative disease like Lou Gehrig's disease. We're able to eavesdrop in on electrical activity and slide a cursor across a keyboard and type out messages. Our research is focused on an area called the "motor cortex."
00:00:34 The motor cortex is responsible for generating movement on the opposite side of the body. Even in cases where you have paralysis, those cells, those neurons in that area of the brain are still active. We can surgically implant a tiny electrical sensor that's made out of silicon. It has 100 tiny electrodes. It just sits on the surface in the brain, and it's able to pick up on the electrical activity
00:01:00 of individual brain cells or neurons. You record tiny signals from those neurons, those brain cells called "action potentials." We can then take those signals and decode them using a computer. We can manipulate them once they're in digital form, like any other data. So all you need to do is imagine moving your right arm, for example, to the letter t on a keyboard, then it'll slide out to the t.
00:01:28 And in addition to that, we're also able to detect when you wish to select that letter t. Our participants were able to type at rates anywhere from 12 to up to almost 40 characters per minute, which translates to, in the best cases, about 6 to 8 words per minute. The interface that you could control with these decoded neural signals is really manyfold. So you might instead wish to control a robotic arm. You could imagine also interfacing with your home--
00:02:02 wirelessly sending signals to your thermostat, open and close doors remotely. This is really the business of Internet of things. It's important to keep in mind that this is a phase one study, so this is really a safety study. It's a feasibility study. It's not meant to be the definitive device that one would deploy. It will obviously be important to have these systems be completely implantable, wireless,
00:02:28 able to function autonomously, not require a technician to set them up. So these are all areas that we're researching throughout our [INAUDIBLE] consortium of which we're a larger part, but I'm very confident that in the not-too-distant future we'll have systems that are deployable and able to provide help for people with paralysis. For more, please visit us at Stanford.edu.
00:02:58