Neural spike compression through salient sample extraction and curve fitting dedicated to high-density brain implants.

Abstract

As brain implants evolve towards higher channel density, efficient on-implant processing of the acquired signals becomes essential to overcome constraints in power, area, and data transmission. Here we propose a data reduction framework, specific to extra-cellular neuronal action potentials. This approach picks a small number of salient spike samples, using which the spike waveshape is interpolated. Attributes of salient samples are sent off the implant to reconstruct the spike waveshape on the external side of the system. In addition to exhibiting high data compression capability, this technique is highly hardware efficient, hence well suits for brain-implantable neural recording microsystems with high channel counts. Based on the proposed framework, a 128-channel neural signal compressor was implemented using a 130-nm CMOS technology, and measured 1.05 × 0.35 mm². At a spike firing rate of 8 Spike/s, the circuit temporally reduces neural data with an average compression rate of ~2176. Operated at 1 V and 32 MHz, the neural data compressor consumes 0.164 µW/channel. The framework proposed in this work substantially reduces the data representing spike waveforms, enabling next-generation, high-density neural recording brain implants to telemeter the acquired neuronal activities to the outside world.