A 2.5-20 kS/s In-Pixel Direct Digitization ECoG Front End With Submillisecond Stimulation Artifact Recovery

Aditi Jain, Eric Fogleman, Paul Botros, Ritwik Vatsyayan, Asish Koruprolu, Corentin Pochet, Andrew M. Bourhis, Zhaoyi Liu, Suhas Chethan, Hanh-Phuc Le, Ian Galton, Shadi A. Dayeh, Drew A. Hall

Abstract: Neural stimulation is used routinely to diagnose and treat neurological disorders. The stimulation artifacts are, however, problematic for closed-loop neuromodulation therapy, which dynamically adjusts the electrical stimulation parameters based on real-time feedback from the recorded neural activity because they can cause saturation or prolonged recovery times in traditional recording front ends. This article presents a per-pixel second-order $\Delta \Sigma $ analog-to-digital converter (ADC) for direct digitization of neural signals, which addresses the stimulation artifact recovery time in voltage-controlled oscillator (VCO)-based quantizers with a fast-recovery, overrange-detecting phase quantizer. The ADC uses a pseudo-virtual ground feedforwarding (PVG FF) technique and a complementary input ${G} _{\text {m}}$ -C filter with per-pixel decimation. It supports four recording modes covering 2.5–20 kS/s through a power-efficient, bandwidth-scalable continuous time $\Delta \Sigma $ modulator. Fabricated in a 180-nm CMOS process, this $300\times 300~\mu $ m2 ADC achieves $\gt 250\times $ faster (0.05–0.4 ms) stimulation artifact recovery time, enabling in-stimulation recording. Recording with artifact tolerance was demonstrated through an in vivo whisker barrel rat experiment.

External IDs:dblp:journals/jssc/JainFBVKPBLCLGDH25