Bio-inspired Hardware Central Pattern Generator (hCPG) as a Therapy for Cardiorespiratory Disease

Abstract

We report on a hCPG device, which is a network of silicon neurons, and its applications to cardiorespiratory therapy. We study the chaotic dynamics of neurons that compete through mutually inhibitory synapses and demonstrate the emergence of multistable behavior. We were able to select the spatio-temporal sequences associated with stable modes of oscillation by imparting different initial conditions with timed current steps mimicking delayed stimuli. We constructed the phase lag maps of the hCPG for various connectivities of the network. We also describe a simpler two neuron hCPG which generates a two-phase rhythmic pattern for vagus nerve stimulation and modulation of heart rate by respiration to obtain artificial respiratory sinus-arrhythmia (RSA).