To reconstruct the motor and sensory functions of human following brain nerve injury, we design a microelectronic neurobridge device as a replacement for the impaired brain nervous tissues. The sensory and motor functions are reconstructed for the paralyzed limbs. Models of this system are established along with the motor nerve signal detection device, signal acquisition device, microelectronic neurobridge testing machine, fast algorithm validation platform and action potential detection and recognition circuit. To solve the problem of low accuracy in detecting action potentials, we propose the core algorithm for the microelectronic neurobridge. This algorithm is based on constraints on amplitude thresholds and differential thresholds, and the time domain features are considered. The type feature vectors are obtained by K-means clustering with higher sensitivity and specificity. Real-time detection and processing of signals are achieved by using real-time fast algorithm. Human hand movement is detected by the double-channel microelectronic neurobridge device. Our findings provide a new solution to restore the motor functions of paralyzed patients.

Paralysis, Brain Nerve Injury, Microelectronic Neurobridge, Action Potential Detection, Motor Function Reconstruction