Tensor Convolutional Sparse Coding with Low-Rank activations, an application to EEG analysis

TL;DR

This paper introduces a novel tensor convolutional sparse coding model leveraging low-rank tensor structures for EEG analysis, improving robustness and interpretability of neural signal representations during anesthesia.

Contribution

It proposes the Kruskal CSC model and an efficient optimization algorithm, TC-FISTA, specifically designed for tensor-based EEG data analysis.

Findings

TC-FISTA effectively handles noisy EEG data.

The model produces sparse, interpretable encodings.

Results demonstrate robustness and accuracy on real EEG data.

Abstract

Recently, there has been growing interest in the analysis of spectrograms of ElectroEncephaloGram (EEG), particularly to study the neural correlates of (un)-consciousness during General Anesthesia (GA). Indeed, it has been shown that order three tensors (channels x frequencies x times) are a natural and useful representation of these signals. However this encoding entails significant difficulties, especially for convolutional sparse coding (CSC) as existing methods do not take advantage of the particularities of tensor representation, such as rank structures, and are vulnerable to the high level of noise and perturbations that are inherent to EEG during medical acts. To address this issue, in this paper we introduce a new CSC model, named Kruskal CSC (K-CSC), that uses the Kruskal decomposition of the activation tensors to leverage the intrinsic low rank nature of these representations in order to extract relevant and interpretable encodings. Our main contribution, TC-FISTA, uses multiple tools to efficiently solve the resulting optimization problem despite the increasing complexity induced by the tensor representation. We then evaluate TC-FISTA on both synthetic dataset and real EEG recorded during GA. The results show that TC-FISTA is robust to noise and perturbations, resulting in accurate, sparse and interpretable encoding of the signals.