Stanford University researchers have managed to use an implanted multi-electrode array to read the brain signals of monkeys thinking of typing letters on a screen and to convert those signals into the target letters at a rate of 12 words per minute.
Medgadget September 13, 2016
Such technology has been tried in the past on humans, but the results have been somewhat underwhelming because the typing was really slow. At 12 words per minute things are looking way more impressive and the research gives a great deal of hope for people with severe paralysis to be able to communicate better, control electronic devices, and regain a voice that they lost to injury or disease.
While the hardware has the capability of reading signals at high resolution, having software interpret them quickly enough was the challenge the team worked on in this latest iteration of the project. The electrodes are positioned over the brain area responsible for hand and arm movement and the users of the system were expected to move the mouse cursor on the screen as though they’re holding onto it with their own hand.
Since monkeys aren’t good at spelling, the interface lights up the letter that needs to be pressed and the animals were taught that clicking along gets them a treat. After a bit of training and refining the data processing algorithms, the team was able to achieve this impressive feat. While impressive in itself, there’s a great deal of room to make the system even faster if a bit of word completion and other technologies we have on our smartphones can be integrated into the software.
Here’s an example of a monkey typing through thought alone:
|Stanford researchers develop technology that allows typing based on brain signals. A team led by electrical engineer Krishna Shenoy developed technology that detects brain signals to move a cursor. Animals trained to copy text using the technology were able to type at a rate of 12 words per minute. Stanford. YouTube Sep 7, 2016|
A Nonhuman Primate Brain–Computer Typing Interface, Paul Nuyujukian, Jonathan C. Kao, Stephen I. Ryu, Krishna V. Shenoy. Proceedings of the IEEE Volume: PP, Issue: 99. Print ISSN: 0018-9219 Online ISSN: 1558-2256DOI: 10.1109/JPROC.2016.2586967
The Science of Neural Interface Systems, Nicholas G. Hatsopoulos and John P. Donoghue. Annu Rev Neurosci. Annu Rev Neurosci. 2009; 32: 249–266. doi: 10.1146/annurev.neuro.051508.135241
Brain-Computer Interfaces in Medicine, Jerry J. Shih, Dean J. Krusienski and Jonathan R. Wolpawc. Mayo Clin Proc. 2012 Mar; 87(3): 268–279. doi: 10.1016/j.mayocp.2011.12.008
ALS Patients Set New Typing World Record Medgadget
Fully Implantable Wireless Brain-Machine Interface Announced Medgadget
Samsung Demos a Tablet Controlled by Your Brain MIT Technology Review
Mind-controlled robot arms show promise Nature
Monkeys transcribe Hamlet with new brain-reading tech Engineering and Technology Magazine
Stephen Hawking’s new Intel talking system to be made open-source Engineering and Technology Magazine