Electrical Stimulus Artifact Cancellation and Neural Spike Detection on Large Multi-Electrode Arrays

Abstract

Simultaneous electrical stimulation and recording using multi-electrode arrays can provide a valuable technique for studying circuit connectivity and engineering neural interfaces. However, interpreting these measurements is challenging because the spike sorting process (identifying and segregating action potentials arising from different neurons) is greatly complicated by electrical stimulation artifacts across the array, which can exhibit complex and nonlinear waveforms, and overlap temporarily with evoked spikes. Here we develop a scalable algorithm based on a structured Gaussian Process model to estimate the artifact and identify evoked spikes. The effectiveness of our methods is demonstrated in both real and simulated 512-electrode recordings in the peripheral primate retina with single-electrode and several types of multi-electrode stimulation. We establish small error rates in the identification of evoked spikes, with a computational complexity that is compatible with real-time data analysis. This technology may be helpful in the design of future high-resolution sensory prostheses based on tailored stimulation (e.g., retinal prostheses), and for closed-loop neural stimulation at a much larger scale than currently possible. Author summarySimultaneous electrical stimulation and recording using multi-electrode arrays can provide a valuable technique for studying circuit connectivity and engineering neural interfaces. However, interpreting these recordings is challenging because the spike sorting process (identifying and segregating action potentials arising from different neurons) is largely stymied by electrical stimulation artifacts across the array, which are typically larger than the signals of interest. We develop a novel computational framework to estimate and subtract away this contaminating artifact, enabling the large-scale analysis of responses of possibly hundreds of cells to tailored stimulation. Importantly, we suggest PLOS Computational Biology | https://doi

An unsupervised real-time spike sorting system based on optimized OSort

“…Another challenge is when multiple spikes from different neurons happen within a narrow time window and are recorded as a single composite waveform, thus requiring more robust sorting algorithms [59] and extra efforts on hardware designs. In addition, the artifact introduced by electrical stimulation and/or motion artifacts during freely moving animals also requires extra preprocessing as artifact removal [60][61][62]. Nevertheless, once the 'processed' spikes have been obtained, the sorting function can still be executed by the proposed opt-OSort.…”

Learning a neural response metric for retinal prosthesis

“…Due to the limited duration of experiments, we focused on stimulation patterns in which only one electrode was active. The data was spike sorted and spiking probability was computed for each cell by averaging across trials for each electrical stimulation pattern (Mena et al, 2016). The probability of firing for each cell in response to increasing current amplitude through each electrode was approximated with a sigmoid function.…”