Brain-computer interface achieves fastest-ever typing results
Wed 22 Feb 2017
A new development in brain-computer interface (BCI) communication could allow people with paralysis to type at the highest speeds and accuracy rates to date.
A research consortium based at Stanford University, known as BrainGate, has built the new interface which could let paralysed people type at a faster rate than ever before possible using only their thoughts.
While BCI technologies have been around since the 70s, few models have been sufficiently responsive for wide-spread rollout. The new BrainGate study hopes to change this.
The clinical trial included three volunteers each suffering from severe limb paralysis disorders – two from amyotrophic lateral sclerosis, or Lou Gehrig’s disease, and one from a spinal cord injury.
The participants agreed to have one or two small ‘aspirin-sized’ electrode arrays implanted into their brains. These silicon chips were then used to record signals from individual nerve cells in the motor cortex, before a series of algorithms translated these into computer commands – in this case, moving a cursor to select desired letters.
Following a brief training period, the volunteers were able to outperform all existing brain-computer interface test results. The study noted one participant, Dennis Degray from Menlo Park, California, was able to achieve 39 correctly-typed characters, the equivalent of around eight words, per minute. The BrainGate team suggested that this could even be boosted with the support of automatic word completion software.
Degray commented on the innovation: “This is like one of the coolest video games I’ve ever gotten to play with. And I don’t even have to put a quarter in it.”
‘This study reports the highest speed and accuracy, by a factor of three, over what’s been shown before,’ said electrical engineering professor and senior author on the study Krishna Shenoy. ‘We’re approaching the speed at which you can type text on your cellphone.’
Looking five years ahead, the researcher hopes that future versions of the equipment will be wireless, fully implanted, self-calibrating and able to provide support 24/7.
‘I don’t see any insurmountable challenges,’ he said. ‘We know the steps we have to take to get there.’