Recent Advances in Brain-controlled Prosthetics for Paralysis: Friday Keynote

Abstract

Neurons encode many parameters simultaneously, but the encoding fidelity at the level of individual neurons is weak. In contrast, with a better understanding of neural population function we can now decode complex arm and hand movement. We have developed a simple extraction algorithm to capture arm movement data and shown that a paralyzed patient who cannot move any part of her body below her neck can control a high-performance "modular prosthetic limb" using 10 degrees-of-freedom simultaneously. The control of this artificial limb is intuitive, with coordinated, graceful motion, closely resembling natural arm and hand movement.