An active micro-electrode array with spike detection and asynchronous readout

Abstract

We present an active micro-electrode array for neural recording with integrated spike detection and an asynchronous readout architecture. Neural amplifier arrays generate voluminous data because of the necessary per-channel sampling rates and number of channels in a dense array. Most of the time, neural cells produce well below 100 spikes per second, with action potential durations generally on the order of 1 ms, and accordingly much of the recorded data from a neural amplifier is not of interest. In the case of dense arrays recording from single units, only the timing of action potentials is relevant and spike sorting is not required. In such a case, the bandwidth requirement of the neural array can be reduced by employing an event-driven data communication protocol such as address event representation (AER). In our array, these events are generated by the spike detection circuits and then relayed to AER modules that send the address of the spiking neuron off-chip using a digital encoding scheme. Based on simulation data, the system implemented here reduces bandwidth requirements by a factor of 1600 in comparison to traditional synchronous sampling.