Learning Robust Features using Deep Learning for Automatic Seizure Detection

TL;DR

This paper introduces a deep neural network that effectively learns robust features from EEG data to automatically detect seizures, handling variability across patients and electrode configurations.

Contribution

The study presents a novel recurrent convolutional neural network that captures spectral, temporal, and spatial EEG features, outperforming previous methods in seizure detection accuracy.

Findings

Significantly improved sensitivity and false positive rates over prior cross-patient classifiers.

Robustness to missing EEG channels and different electrode montages.

Effective generalization across diverse patient data.

Abstract

We present and evaluate the capacity of a deep neural network to learn robust features from EEG to automatically detect seizures. This is a challenging problem because seizure manifestations on EEG are extremely variable both inter- and intra-patient. By simultaneously capturing spectral, temporal and spatial information our recurrent convolutional neural network learns a general spatially invariant representation of a seizure. The proposed approach exceeds significantly previous results obtained on cross-patient classifiers both in terms of sensitivity and false positive rate. Furthermore, our model proves to be robust to missing channel and variable electrode montage.