ATR – DNI

Patients with ALS or stroke, sometimes go into a total locked-in state. In this state, they can not move their muscles at all even when they are clearly thinking what they want to do. Brain-machine interface, or BMI, is one of the ways to improve their quality of life. The BMI infers the patient’s will only using their brain signals and translates them to control some external devices, such as prosthetic hand.
In this study, we succeeded to develop an integrated system to control a prosthetic hand using human electrocorticograms in real-time. Notably, this new system was applied to a post-stroke patient with surprisingly good results. We found that the patient’s hand movements were correctly detected and inferred only by using the electrocroticographic signals recorded on the patient’s sensorimotor cortex. Actually, the prosthetic hand was able to mimic the patient’s movements in real-time. This is the first report describing the use of the electrocorticographic signals to control the prosthetic hand. Because the signals are known to be stable and durable signals for the clinically feasible BMI systems, this success paves the way for the motor restoration of paralyzed patients with using the BMI in daily clinical practice.