Automated Detection of Epileptic Spikes and Seizures Incorporating a Novel Spatial Clustering Prior

TL;DR

This paper presents a novel spatial clustering prior and a specialized neural network architecture for automated epileptic spike and seizure detection in MEG and EEG data, achieving state-of-the-art accuracy and stability.

Contribution

It introduces a spatial clustering approach combined with a new convolutional module and a custom ResNet3D model for improved detection accuracy.

Findings

Achieved 94.73% F1 score on MEG dataset, outperforming previous methods.

Demonstrated improved weighted F1 score of 1.4% on EEG seizure detection.

Effective in both MEG spike detection and EEG seizure classification.

Abstract

A Magnetoencephalography (MEG) time-series recording consists of multi-channel signals collected by superconducting sensors, with each signal's intensity reflecting magnetic field changes over time at the sensor location. Automating epileptic MEG spike detection significantly reduces manual assessment time and effort, yielding substantial clinical benefits. Existing research addresses MEG spike detection by encoding neural network inputs with signals from all channel within a time segment, followed by classification. However, these methods overlook simultaneous spiking occurred from nearby sensors. We introduce a simple yet effective paradigm that first clusters MEG channels based on their sensor's spatial position. Next, a novel convolutional input module is designed to integrate the spatial clustering and temporal changes of the signals. This module is fed into a custom MEEG-ResNet3D developed by the authors, which learns to extract relevant features and classify the input as a spike clip or not. Our method achieves an F1 score of 94.73% on a large real-world MEG dataset Sanbo-CMR collected from two centers, outperforming state-of-the-art approaches by 1.85%. Moreover, it demonstrates efficacy and stability in the Electroencephalographic (EEG) seizure detection task, yielding an improved weighted F1 score of 1.4% compared to current state-of-the-art techniques evaluated on TUSZ, whch is the largest EEG seizure dataset.