HermEIS: A Parallel Multichannel Approach to Rapid Spectral Characterization of Neural MEAs

Abstract

The promise of increasing channel counts in high density ($>10^4$) neural Microelectrode Arrays (MEAs) for high resolution recording comes with the curse of developing faster characterization strategies for concurrent acquisition of multichannel electrode integrities over a wide frequency spectrum. To circumvent the latency associated with the current multiplexed technique for impedance acquisition, it is common practice to resort to the single frequency impedance measurement (i.e. $Z_{1\text{kHz}}$). This, however, does not offer sufficient spectral impedance information crucial for determining the capacity of electrodes at withstanding slow and fast-changing stimulus and recordings. In this work, we present \textit{HermEIS}, a novel approach that leverages single cycle in-phase and quadrature signal integrations for reducing the massive data throughput characteristic of such high density acquisition systems. As an initial proof-of-concept, we demonstrate over $6$ decades of impedance bandwidth ($5\times 10^{-2}-5\times 10^4\text{ Hz}$) in a parallel $4$-channel potentiostatic setup composed of a custom PCB with off-the-shelf electronics working in tandem with an FPGA.

5 pages, submitted to IEEE EMBC 2024