A Novel Independent RNN Approach to Classification of Seizures against Non-seizures

TL;DR

This paper introduces a new deep learning approach using IndRNNs for classifying seizures from EEG data, effectively capturing temporal features across various time scales and outperforming existing methods.

Contribution

The study develops an IndRNN-based method that expands time scales for improved seizure detection, addressing variability in seizure morphologies and emphasizing the importance of segment length.

Findings

Outperforms current state-of-the-art methods in seizure classification

Segment length significantly impacts classification accuracy

Maximum performance variation exceeds 4% across different segment lengths

Abstract

In current clinical practices, electroencephalograms (EEG) are reviewed and analyzed by trained neurologists to provide supports for therapeutic decisions. Manual reviews can be laborious and error prone. Automatic and accurate seizure/non-seizure classification methods are desirable. A critical challenge is that seizure morphologies exhibit considerable variabilities. In order to capture essential seizure features, this paper leverages an emerging deep learning model, the independently recurrent neural network (IndRNN), to construct a new approach for the seizure/non-seizure classification. This new approach gradually expands the time scales, thereby extracting temporal and spatial features from the local time duration to the entire record. Evaluations are conducted with cross-validation experiments across subjects over the noisy data of CHB-MIT. Experimental results demonstrate that the proposed approach outperforms the current state-of-the-art methods. In addition, we explore how the segment length affects the classification performance. Thirteen different segment lengths are assessed, showing that the classification performance varies over the segment lengths, and the maximal fluctuating margin is more than 4%. Thus, the segment length is an important factor influencing the classification performance.