Paralysis following spinal cord injury, brainstemstroke, amyotrophic\udlateral sclerosis and other disorders can disconnect the brain from the\udbody, eliminating the ability to perform volitional movements. A\udneural interface system could restore mobility and independence\udfor people with paralysis by translating neuronal activity directly into\udcontrol signals for assistive devices. We have previously shown that\udpeople with long-standing tetraplegia can use a neural interface\udsystem to move and click a computer cursor and to control physical\uddevices. Able-bodied monkeys have used a neural interface system\udto control a robotic arm, but it is unknown whether people with\udprofound upper extremity paralysis or limb loss could use cortical\udneuronal ensemble signals to direct useful arm actions. Here we\uddemonstrate the ability of two people with long-standing tetraplegia\udto use neural interface system-based control of a robotic arm to\udperform three-dimensional reach and graspmovements. Participants\udcontrolled the arm and hand over a broad space without explicit\udtraining, using signals decoded from a small, local population of\udmotor cortex (MI) neurons recorded from a 96-channel microelectrode\udarray. One of the study participants, implanted with the\udsensor 5 years earlier, also used a robotic arm to drink coffee from a\udbottle. Although robotic reach and grasp actions were not as fast or\udaccurate as those of an able-bodied person, our results demonstrate\udthe feasibility for people with tetraplegia, years after injury to the\udcentral nervous system, to recreate useful multidimensional control\udof complex devices directly from a small sample of neural signals.
展开▼
