Multivariate temporal dictionary learning for EEG

Abstract

This article addresses the issue of representing electroencephalographic (EEG) signals in an efficient way. While classical approaches use a fixed Gabor dictionary to analyze EEG signals, this article proposes a data-driven method to obtain an adapted dictionary. To reach an efficient dictionary learning, appropriate spatial and temporal modeling is required. Inter-channels links are taken into account in the spatial multivariate model, and shift-invariance is used for the temporal model. Multivariate learned kernels are informative (a few atoms code plentiful energy) and interpretable (the atoms can have a physiological meaning). Using real EEG data, the proposed method is shown to outperform the classical multichannel matching pursuit used with a Gabor dictionary, as measured by the representative power of the learned dictionary and its spatial flexibility. Moreover, dictionary learning can capture interpretable patterns: this ability is illustrated on real data, learning a P300 evoked potential.

Analytical methods and experimental approaches for electrophysiological studies of brain oscillations

“…A possible drawback lies in the fact that MP algorithms consist of a finite set of atoms and therefore introduce an estimation bias towards these atoms (although this can be addressed with stochastic dictionaries (Durka et al, 2001) or dictionary learning (Barthélemy et al, 2013)).…”

Quantitative Comparison of EEG Compressed Sensing using Gabor and K-SVD Dictionaries

“…Fauvel and Rabab [24] noted that the Basis pursuit de-noise algorithm (BPDN) required fewer measurements than other greedy algorithms to achieve a comparable reconstruction quantity. Moreover, BPDN is packed in Matlab solver for a large-scale one-norm regularized least squares named SPGL1 [25]. For these reasons, we selected BPDN as our reconstruction algorithm.…”

Markov Chain Monte Carlo Inference of Parametric Dictionaries for Sparse Bayesian Approximations

“…Other efforts have proposed Gaussian-like functions to efficiently capture spherical stereo images [70], diffusion-based dictionaries to model MRI [25], and other wavelet-like atoms for digitizing fingerprint images [71]. Gabor dictionaries have been used for the electroencephalogram (EEG) [72], spline wavelets for the electrocardiogram (ECG) [73], and sigmoid-exponential functions for the electrodermal activity (EDA) [28]. …”

“…Parametric DL is more likely to converge faster and have more efficient implementations compared to the non-parametric problem [19]. It further provides higher signal interpretability yielding important metainformation [25], [26], [27], [28]. …”