Validating the signal fidelity of novel soft neural probes

This project, published in Nature Communications, aims to address the challenges of integrating soft neural probes with biological tissues. By utilizing high-resolution printing of liquid electronics, we can create interconnects directly on the cranium, minimizing mechanical mismatch and improving long-term stability.

As the sole statistician on the team, I applied signal processing and statistical testing to detect correlations between individual- and population-level neural responses to visual stimuli, thereby validating the signal fidelity of the novel neural probe. Additionally, I built an object-oriented MATLAB data pipeline (github.com/Jong-Min-Moon/neurosignal-matlab) for signal processing, feature extraction, and analysis of neural time-series data.

This page explains core concepts used in this project, including convolution, Fourier transform, Hilbert transform, and wavelet transform, and how they are used to analyze neural data in this paper.