Automatic Detection of Epileptic EEG Signals Based on Channel Attention and Bidirectional Temporal Modeling
Main Article Content
Keywords
epileptic seizure detection, EEG signals, deep learning, channel attention, bidirectional long short-term memory (BiLSTM), one-dimensional convolutional neural network (1D-CNN)
Abstract
Epilepsy is a common chronic neurological disease. As an important tool for epilepsy diagnosis and disease assessment, electroencephalogram (EEG) can reflect the abnormal discharge activity of brain neurons. However, traditional EEG interpretation relies heavily on expert manual analysis, which has problems such as time-consuming, strong subjectivity and low efficiency. To improve the automation level of seizure detection, this paper proposes a pyramid-type one-dimensional convolutional neural network model (SE-P1D-LSTM) that integrates a channel attention mechanism and a Bidirectional Long Short-Term Memory (BiLSTM) network. The method first preprocesses the EEG signal through overlapping sliding window and Z-Score standardization, and combines Gaussian noise disturbance and random amplitude scaling for data augmentation; Then use the pyramidal one-dimensional convolutional structure to extract local temporal features, and adaptively strengthen the key channel information through SEBlock; Finally, a BiLSTM is introduced to model the long-range temporal dependencies in EEG signals to realize the automatic classification of normal and seizure EEG signals. The experiment was carried out based on the public epilepsy EEG dataset of the University of Bonn in Germany, and the performance of the model was evaluated using 10-fold cross-verification. The results show that the method in this paper has achieved an accuracy rate of 99.87%, a sensitivity of 99.87% and a specificity of 99.87% in the two-classification task, which is better than a variety of comparative models. The research results show that SE-P1D-LSTM can effectively extract the discriminative characteristics of epilepsy EEG signals, and has good classification performance and application potential in the automatic detection task of seizures.
References
- [1] GBD 2021 Epilepsy Collaborators. (2025). Global, regional, and national burden of epilepsy, 1990–2021: A systematic analysis for the Global Burden of Disease Study 2021. The Lancet Neurology.
- [2] Guo, J., & Wei, X. N. (2007). Research progress in the application of electroencephalography. Journal of Continuing Education of Shaanxi Normal University, 24(2), 122–124.
- [3] Li, Y. (2025). Research on epileptic EEG signal analysis methods based on graph neural networks (Master’s thesis, Shandong University, Jinan, China).
- [4] Subasi, A., & Ercelebi, E. (2005). Classification of EEG signals using neural network and logistic regression. Computer Methods and Programs in Biomedicine, 78(2), 87–99.
- [5] Guo, L., Rivero, D., Dorado, J., et al. (2010). Automatic epileptic seizure detection in EEGs based on line length feature and artificial neural networks. Journal of Neuroscience Methods, 191(1), 101–109.
- [6] Alotaiby, T. N., Alshebeili, S. A., Alshawi, T., et al. (2014). EEG seizure detection based on linear and nonlinear features using artificial neural networks. Clinical Neurophysiology, 125(7), 1405–1414.
- [7] Acharya, U. R., Fujita, H., Lih, O. S., et al. (2017). Automated detection of seizures using a deep convolutional neural network. International Journal of Environmental Research and Public Health, 14(10), 1182.
- [8] Tsiouris, K. M., Tzallas, A. T., Tsipouras, M. G., et al. (2018). A long short-term memory deep learning network for the prediction of epileptic seizures using EEG signals. Computers in Biology and Medicine, 99, 24–37.
- [9] Rukhsar, S., & Tiwari, A. K. (2024). Lightweight convolution transformer for cross-patient seizure detection in multi-channel EEG signals. Biomedical Signal Processing and Control, 95, 106452.
- [10] Li, X. S., Hu, C. X., Tian, Z. J., et al. (2026). Lightweight attention residual-enhanced YOLO architecture for remote sensing object detection. Journal of Geo-Information Science.
- [11] Wang, W. H. (2024). Semi-supervised ship engine state recognition method based on efficient self-attention (Master’s thesis, Dalian Maritime University, Dalian, China).
- [12] Zhang, W. S. (2025). Deep learning-based network intrusion detection methods (Master’s thesis, Guilin University of Electronic Technology, Guilin, China).
- [13] Luo, J. (2025). Chinese text classification based on syntactic analysis and abstract semantic representation (Master’s thesis, Wuhan Institute of Technology, Wuhan, China).
- [14] Li, J. H., Gao, L. F., & Li, X. W. (2026). Application of surface acoustic wave wireless temperature sensors in fault prediction of power equipment. Electric Power Equipment Management.
- [15] Chen, N. (2025). Research on molecular property prediction based on multi-view graph learning (Master’s thesis, Nanning Normal University, Nanning, China).
- [16] Ruan, Z. J. (2025). Research on sequence recommendation algorithms for bi-objective optimization (Master’s thesis, Guangdong University of Foreign Studies, Guangzhou, China).
- [17] Li, X. L., Zhang, Y., Zhang, W., et al. (2024). State-of-health prediction for lithium-ion batteries based on health factors and CNN-BiLSTM. Journal of Nanjing University of Information Science & Technology.
- [18] Huang, Q. M. (2024). River water level prediction method based on long short-term memory networks. Information Recording Materials.
- [19] Andrzejak, R. G., Lehnertz, K., Morrison, C., et al. (2001). Indications of nonlinear deterministic and finite-dimensional structures in time series of brain electrical activity: Dependence on recording region and brain state. Physical Review E, 64(6), 061907.
- [20] Wang, J., & Li, L. Y. (2025). Breast ultrasound lesion segmentation network based on channel-wise spatial adaptive selective kernel convolution and bidirectional boundary perception mechanism. Journal of South China University of Technology (Natural Science Edition).
- [21] Zhang, X. T. (2025). Application of deep learning algorithms integrating Transformer and mixture-of-experts in sleep staging (Master’s thesis, Shanghai Polytechnic University, Shanghai, China).
- [22] Torse, D. A., & Khanai, R. (2021). Classification of epileptic seizures using ensemble empirical mode decomposition and least squares support vector machine. In 2021 International Conference on Computer Communication and Informatics (ICCCI) (pp. 1–5). IEEE.
- [23] Rashed-Al-Mahfuz, M., et al. (2021). A deep convolutional neural network method to detect seizures and characteristic frequencies using epileptic electroencephalogram (EEG) data. IEEE Access, 9, 1–12.
- [24] Liu, X., Wang, J., Shang, J., et al. (2022). Epileptic seizure detection based on variational mode decomposition and deep forest using EEG signals. Biosensors, 12(10), 1275.
- [25] Huang, L., Zhou, K., Chen, S., et al. (2024). Automatic detection of epilepsy from EEGs using a temporal convolutional network with a self-attention layer. Bioengineering, 23(1), 50.
- [26] Jain, M., Bhardwaj, H., & Srivastav, A. (2024). Bayesian-enhanced EEG signal analysis for epileptic seizure recognition: A 1D-CNN LSTM approach. In 2024 11th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO) (pp. 1–6). IEEE.
Article Sidebar
This work is licensed under a Creative Commons Attribution 4.0 International License.