MR-EEGWaveNet: Multiresolutional EEGWaveNet for Seizure Detection from Long EEG Recordings

TL;DR

This paper introduces MR-EEGWaveNet, an end-to-end multiresolutional model that improves seizure detection accuracy from long EEG recordings by capturing temporal and spatial features and reducing false positives.

Contribution

The study presents a novel multiresolutional EEGWaveNet model with an anomaly score-based post-processing technique, outperforming traditional methods in seizure detection accuracy.

Findings

F1 score improved from 0.177 to 0.336 on Siena dataset.

F1 score improved from 0.327 to 0.488 on Juntendo dataset.

Model significantly reduces false positives.

Abstract

Feature engineering for generalized seizure detection models remains a significant challenge. Recently proposed models show variable performance depending on the training data and remain ineffective at accurately distinguishing artifacts from seizure data. In this study, we propose a novel end-to-end model, "Multiresolutional EEGWaveNet (MR-EEGWaveNet)," which efficiently distinguishes seizure events from background electroencephalogram (EEG) and artifacts/noise by capturing both temporal dependencies across different time frames and spatial relationships between channels. The model has three modules: convolution, feature extraction, and predictor. The convolution module extracts features through depth-wise and spatio-temporal convolution. The feature extraction module individually reduces the feature dimension extracted from EEG segments and their sub-segments. Subsequently, the extracted features are concatenated into a single vector for classification using a fully connected classifier called the predictor module. In addition, an anomaly score-based post-classification processing technique is introduced to reduce the false-positive rates of the model. Experimental results are reported and analyzed using different parameter settings and datasets (Siena (public) and Juntendo (private)). The proposed MR-EEGWaveNet significantly outperformed the conventional non-multiresolution approach, improving the F1 scores from 0.177 to 0.336 on Siena and 0.327 to 0.488 on Juntendo, with precision gains of 15.9% and 20.62%, respectively.