GRC-Net: Gram Residual Co-attention Net for epilepsy prediction

CGI2018 manuscript No. (will be inserted by the editor) GRC-Net: Gram Residual Co-attention Net for epilepsy prediction Bihao You · Jiping Cui Abstract Prediction of epilepsy based on electroen- cephalogram (EEG) signals is a rapidly evolving field. Previous studies have traditionally applied 1D process- ing to the entire EEG signal. However, we have adopted the Gram Matrix method to transform the signals into a 3D representation, enabling modeling of signal relation- ships across dimensions while preserving the temporal dependencies of the one-dimensional signals. Addition- ally, we observed an imbalance between local and global signals within the EEG data. Therefore, we introduced multi-level feature extraction, utilizing coattention for capturing global signal characteristics and an inception structure for processing local signals, achieving multi- granular feature extraction. Our experiments on the BONN dataset demonstrate that for the most challeng- ing five-class classification task, GRC-Net achieved an accuracy of 93.66%, outperforming existing methods. Keywords epilepsy prediction · Signal Process- ing · Gramian Angular Feild · Multi-level Feature Extraction · Cotattention · Computational Perception 1 Introduction EEG signals are indispensable for analyzing brain ac- tivity, understanding brain function, and exploring cog- nitive mechanisms. Therefore, effective processing of Bihao You School of Advanced Technology, Xi’an Jiaotong-Liverpool University, Suzhou, 215123, China E-mail: Bihao.You22@student.xjtlu.edu.cn Jiping Cui School of Advanced Technology, Xi’an Jiaotong-Liverpool University, Suzhou, 215123, China; and also with the School of Electrical Engineering, Electron- ics and Computer Science, University of Liverpool, Liverpool, L69 3BX, United Kingdom E-mail: Jiping.Cui24@student.xjtlu.edu.cn EEG signals is crucial in various domains such as speech recognition [5], emotion detection [20], and diagnosing diseases related to brain function [3] [18] [8]. While sig- nificant progress has been made in using one-dimensional EEG signal processing for predicting epileptic seizures, there is a growing interest in transforming one-dimensional EEG signal sequences into two-dimensional images to extract deep features from multiple channels for predic- tion. Fig. 1 In the first column of images, the original one- dimensional EEG signals from the BoNN dataset are dis- played. In the second column, the signals underwent resam- pling using sliding windows with specified strides. In the third column, the resampled signals were transformed into feature maps using the Gram Matrix method for standardization, while preserving temporal dependencies. With the emergence of EEG datasets for epileptic analysis, such as BONN, several methods have been proposed for seizure prediction. Although these works identified some issues in using EEG signal processing for seizure prediction, their models addressed them in a rudimentary manner. Previous studies on EEG signal prediction have mostly relied on one-dimensional time series or simply utilized arXiv:2512.12273v1 [cs.LG] 13 Dec 2025 2 Bihao You, Jiping Cui time series to plot corresponding curves. However, we argue that due to the unidirectional correlation in the horizontal and vertical directions of one-dimensional signals, they may not fully capture the underlying rela- tionships in the data. Moreover, the linear operations on one-dimensional sequences fail to discriminate between the meaningful information in the signal and Gaussian noise. Therefore, the utilization of one-dimensional sig- nals alone for plotting corresponding images may lead to a loss of temporal dependencies in the data, resulting in information loss. Furthermore,traditional CNN-based models lack the ability to model long-range dependencies and percep- tions due to their focus on modeling local informa- tion. While some researchers have recognized this is- sue and attempted to incorporate attention mechanisms in models for long-range global modeling, the original Self-Attention structure in Transformers calculates at- tention matrices based on interactions between queries and keys, thereby overlooking the relationships between adjacent keys. This limitation still hampers the effec- tiveness of global modeling. To tackle the primary concern, we propose a tech- nique that converts the original EEG signal based on Gramian Angular Feild(GAF). The GAF method is em- ployed to map one-dimensional EEG signal data onto a polar coordinate system, where as the sequence pro- gresses over time, the corresponding values are distorted between different angular points on the circle. This trans- formation generates a bidirectional sequence that re- tains all information without any loss and preserves temporal dependencies. Specifically,as shown in Fig1,we segment the raw data using a sliding window with a specific step size, followed by normalization and sig- nal transformation. Furthermore, to fully leverage the

This content is AI-processed based on open access ArXiv data.