|
|
|
| EEG Signal Classificaiton Method Based on Fuzzy Confidence Causal Power and Multi-stream Ensemble Neural Networks |
| FAN Min1,2,3, CHEN Qiuyu1,2,3, LI Jinhai1,2,3 |
1. Data Science Research Center, Kunming University of Science and Technology, Kunming 650500;
2. Faculty of Science, Kunming University of Science and Technology, Kunming 650500;
3. Yunnan Key Laboratory of Complex Systems and Brain-Inspired Intelligence, Kunming University of Science and Technology, Kunming 650500 |
|
|
|
|
Abstract The automatic identification of epileptic electroencephalogram（EEG） signals is of paramount importance for clinical diagnosis. However, high-dimensional redundancy, non-stationarity, and difficulties in cross-scenario generalization are commonly encountered in automatic EEG detection. To address these issues, a EEG signal classificaiton method based on fuzzy confident causal power and multi-stream ensemble neural networks is proposed. First, a feature selection algorithm based on Rényi entropy and fuzzy confident causal power（FCCP） is proposed to mitigate high-dimensional redundancy. Fuzzy Rényi entropy and the theory of confident causal power are introduced to quantify the bidirectional causal effects between features and labels. Key features with strong causal interpretability are selected by utilizing data structure constraints. Second, the formal context and the space of causal power attribute network are constructed to mine underlying topological features. Thus, network-based modeling of unstructured data is achieved. Finally, a multi-stream ensemble neural network（MENN） classifier is developed. The deep fully-connected stream, the wide stream and the narrow-deep stream are integrated to simultaneously capture local details and global dependencies. Multi-scale feature representations are fused through a concatenation layer. The experiments on Bonn and CHB-MIT clinical datasets demonstrate that the proposed method achieves stable and superior performance under multi-task settings, exhibiting significant robustness in distinguishing seizure signals. The effectiveness of the combination of confident causal power-based feature selection with multi-stream integrated learning is verified.
|
|
Received: 09 February 2026
|
|
|
| Fund:National Natural Science Foundation of China（No.62476114）, Yunnan Fundamental Research Project（No.202401AV070009） |
|
Corresponding Authors:
LI Jinhai, Ph.D., professor. His research interests include cognitive computing, granular computing, big data analysis, concept lattice and rough set.
|
| About author:: About Author:FAN Min, Ph.D., associate professor. Her research interests include data mining, rough set, granular computing and social network analysis. CHEN Qiuyu, Master student. His research interests include network formal context and causal power. |
|
|
|
[1] TRINKA E, COCK H, HESDORFFER D, et al. A Definition and Classification of Status Epilepticus-Report of the ILAE Task Force on Classification of Status Epilepticus. Epilepsia, 2015, 56(10): 1515-1523.
[2] FEI K L, WANG W, YANG Q L, et al. Chaos Feature Study in Fractional Fourier Domain for Preictal Prediction of Epileptic Seizure. Neurocomputing, 2017, 249(C): 290-298.
[3] WANG C S, YI H, WANG W, et al. Lesion Localization Algorithm of High-Frequency Epileptic Signal Based on Teager Energy Operator. Biomedical Signal Processing and Control, 2019, 47: 262-275.
[4] ZHANG J C, ZHENG S J, CHEN W N, et al. A Scheme Combining Feature Fusion and Hybrid Deep Learning Models for Epileptic Seizure Detection and Prediction. Scientific Reports, 2024, 14(1). DOI: 10.1038/s41598-024-67855-4.
[5] HARPALE V, BAIRAGI V. An Adaptive Method for Feature Selection and Extraction for Classification of Epileptic EEG Signal in Significant States. Journal of King Saud University-Computer and Information Sciences, 2021, 33(6): 668-676.
[6] KAYA D. The mRMR-CNN Based Influential Support Decision System Approach to Classify EEG Signals. Measurement, 2020, 156. DOI: 10.1016/j.measurement.2020.107602.
[7] KHOSLA A, KHANDNOR P, CHAND T. EEG-Based Automatic Multi-class Classification of Epileptic Seizure Types Using Recurrence Plots. Expert Systems, 2022, 39(5). DOI: 10.1111/exsy.12923.
[8] OMIDVAR M, ZAHEDI A, BAKHSHI H. EEG Signal Processing for Epilepsy Seizure Detection Using 5-Level DB4 Discrete Wavelet Transform, GA-Based Feature Selection and ANN/SVM Classifiers. Journal of Ambient Intelligence and Humanized Computing, 2021, 12(11): 10395-10403.
[9] HUSSEIN R, WANG Z J, WARD R. Ll-Regularization Based EEG Feature Learning for Detecting Epileptic Seizure // Proc of the IEEE Global Conference on Signal and Information Processing. Washington, USA: IEEE, 2016: 1171-1175.
[10] PENG G C, NOURANI M, HARVEY J, et al. Personalized EEG Feature Selection for Low-Complexity Seizure Monitoring. International Journal of Neural Systems, 2021, 31(8). DOI: 10.1142/S0129065721500180.
[11] RIVERA M J, SANCHIS J, CORCHO O, et al. Evaluating CNN Methods for Epileptic Seizure Type Classification Using EEG Data. IEEE Access, 2024, 12: 75483-75495.
[12] OMAR A, ABD EL-HAFEEZ T. Optimizing Epileptic Seizure Re-cognition Performance with Feature Scaling and Dropout Layers. Neural Computing and Applications, 2024, 36(6): 2835-2852.
[13] PEARL J. Causality. Cambridge, UK: Cambridge University Press, 2009.
[14] HOEL E P, ALBANTAKIS L, TONONI G. Quantifying Causal Emer-gence Shows That Macro Can Beat Micro. PNAS, 2013, 110(49): 19790-19795.
[15] VARLEY T F, HOEL E. Emergence as the Conversion of Information: A Unifying Theory. Philosophical Transactions of the Royal Society A, 2022, 380. DOI: 10.1098/rsta.2021.0150.
[16] ZHANG T, SHAN H R, LITTLE M A. Causal GraphSAGE: A Robust Graph Method for Classification Based on Causal Sampling. Pattern Recognition, 2022, 128. DOI: 10.1016/j.patcog.2022.108696.
[17] SUI Y D, WANG X, WU J C, et al. Causal Attention for Interpretable and Generalizable Graph Classification // Proc of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mi-ning. New York, USA: ACM, 2022: 1696-1705.
[18] FAN M, LI H. Graph Classification Method Based on Fuzzy Entropy Causality in a Network Formal Context. IEEE Transactions on Fuzzy Systems, 2024, 32(9): 5018-5032.
[19] YUE X D, ZHOU J, YAO Y Y, et al. Shadowed Neighborhoods Based on Fuzzy Rough Transformation for Three-Way Classification. IEEE Transactions on Fuzzy Systems, 2020, 28(5): 978-991.
[20] ZHANG Y, YAO J T. Game Theoretic Approach to Shadowed Sets: A Three-Way Tradeoff Perspective. Information Sciences, 2020, 507: 540-552.
[21] WANG Z H, CHEN H M, YUAN Z, et al. Multi-scale Fuzzy Entropy-Based Feature Selection. IEEE Transactions on Fuzzy Systems, 2023, 31(9): 3248-3262.
[22] WILLE R. Restructuring Lattice Theory: An Approach Based on Hierarchies of Concepts // Proc of the 7th International Conference on Formal Concept Analysis. Berlin, Germany: Springer, 2009: 314-339.
[23] AHMED M I B, ZAGHDOUD R A, AL-ABDULQADER M, et al. Ensemble Machine Learning Based Identification of Adult Epilepsy. Mathematical Modelling of Engineering Problems, 2023, 10(1): 84-92.
[24] KAPOOR B, NAGPAL B, JAIN P K, et al. Epileptic Seizure Prediction Based on Hybrid Seek Optimization Tuned Ensemble Classifier Using EEG Signals. Sensors, 2022, 23(1). DOI: 10.3390/s23010423.
[25] SUNARYONO D, SARNO R, SISWANTORO J, et al. Hybrid One-Dimensional CNN and DNN Model for Classification Epileptic Seizure. International Journal of Intelligent Engineering and Systems, 2022, 15(6): 492-502.
[26] RENNER R, WOLF S. Smooth Rényi Entropy and Applications // Proc of the International Symposium on Information Theory. Wa-shington, USA: IEEE, 2004: 232.
[27] LAKSHMINARAYANAN B, PRITZEL A, BLUNDELL C. Simple and Scalable Predictive Uncertainty Estimation Using Deep Ensembles // Proc of the 31st International Conference on Neural Information Processing Systems. Cambridge, USA: MIT Press, 2017: 6405-6416.
[28] ZHANG Z R, WU W M, SUN C, et al. Seizure Detection via Deterministic Learning Feature Extraction. Pattern Recognition, 2024, 153. DOI: 10.1016/j.patcog.2024.110466.
[29] SHI C L, GAO J, YU J, et al. A Novel Similarity-Constrained Feature Selection Method for Epilepsy Detection via EEG Signals. Journal of King Saud University Computer and Information Sciences, 2025, 37(6). DOI: 10.1007/s44443-025-00152-w.
[30] LIU S, WANG J, LI S S, et al. Epileptic Seizure Detection and Prediction in EEGs Using Power Spectra Density Parameterization. IEEE Transactions on Neural Systems and Rehabilitation Enginee-ring, 2023, 31: 3884-3894.
[31] AHMAD I, YAO C, LI L, et al. An Efficient Feature Selection and Explainable Classification Method for EEG-Based Epileptic Seizure Detection. Journal of Information Security and Applications, 2024, 80. DOI: 10.1016/j.jisa.2023.103654.
[32] ABUALHASSAN Z, AHMAD I, ABED S, et al. Epileptic Seizure Detection by Learning EEG as Images Using Vision Transformers. Journal of Engineering Research, 2026, 14(1): 936-952.
[33] XIANG F, LIU M Y, CHEN W N, et al. A Deep Hybrid CSAE-GRU Framework with Two-Stage Balancing for Automatic Epileptic Seizure Detection Using EEG-Derived Features. Frontiers in Neuroscience, 2025, 19. DOI: 10.3389/fnins.2025.1698960.
[34] CAO X S, ZHENG S J, ZHANG J C, et al. A Hybrid CNN-Bi-LSTM Model with Feature Fusion for Accurate Epilepsy Seizure Detection. BMC Medical Informatics and Decision Making, 2025, 25(1). DOI: 10.1186/s12911-024-02845-0.
[35] ZHAO W, WANG W F, PATNAIK L M, et al. Residual and Bidirectional LSTM for Epileptic Seizure Detection. Frontiers in Computational Neuroscience, 2024, 18. DOI: 10.3389/fncom.2024.1415967.
[36] ZENG W, ZHANG M L, SHAN L M, et al. Interpretable Classification of Epileptic EEG Signals Using ALIF Decomposition and Attention-Augmented Cascaded Deep Neural Networks. Applied Soft Computing, 2025. DOI: 10.1016/j.asoc.2025.113211.
[37] TANDRA N, AKHTAR N, PADMANABAN K, et al. A Finite-Element Dual-Level Contextual Informed Neural Network with Swarm Space Hopping Algorithm Based Optimal Feature Selection and Detection for EEG-Based Epileptic Seizure Detection. Swarm and Evo-lutionary Computation, 2025, 97. DOI: 10.1016/j.swevo.2025.102072.
[38] QIN J H, LIU Z J, ZHUANG J H, et al. Dual-Modality Transformer with Time Series Imaging for Robust Epileptic Seizure Prediction. Applied Sciences, 2025, 15(3). DOI: 10.3390/app15031538.
[39] ZHAO Y L, KANG Y F, LI W J, et al. Multimodal Feature Re-presentation and Enhancement EEG Signal Processing for Epilepsy Detection. Measurement, 2026, 260. DOI: 10.1016/j.measurement.2025.119877.
[40] LI R X, ZHAO G X, MUIR D R, et al. Real-Time Sub-Milliwatt Epilepsy Detection Implemented on a Spiking Neural Network Edge Inference Processor. Computers in Biology and Medicine, 2024, 183. DOI: 10.1016/j.compbiomed.2024.109225.
[41] ABDULWAHHAB A H, ABDULAAL A H, AL-GHRAIRI A H T, et al. Detection of Epileptic Seizure Using EEG Signals Analysis Based on Deep Learning Techniques. Chaos, Solitons & Fractals, 2024, 181. DOI: 10.1016/j.chaos.2024.114700.
[42] PATEL J. SeizureSight: 2D CNN-LSTM Hybrid for EEG-Based Seizure Prediction // Proc of the 3rd International Conference on Applied Artificial Intelligence and Computing. Washington, USA: IEEE, 2024: 252-256.
[43] DONG X C, WEN Y M, JI D Z, et al. Epileptic Seizure Detection with an End-to-End Temporal Convolutional Network and Bidirectional Long Short-Term Memory Model. International Journal of Neural Systems, 2024, 34(3). DOI: 10.1142/S0129065724500126.
[44] SUN G Z, ZHANG J B, MA Y J, et al. MCCT-Net: A Multi-perspective Convolution and Cross-Connection Transformer Network for Epilepsy Detection. Neurocomputing, 2026. DOI: 10.1016/j.neucom.2025.132043.
[45] HUANG L, ZHOU K Y, CHEN S Y, et al. Automatic Detection of Epilepsy from EEGs Using a Temporal Convolutional Network with a Self-Attention Layer. Biomedical Engineering Online, 2024, 23(1). DOI: 10.1186/s12938-024-01244-w.
[46] SONG Y B, FAN C L, MAO X Q. Optimization of Epilepsy Detection Method Based on Dynamic EEG Channel Screening. Neural Networks, 2024, 172. DOI: 10.1016/j.neunet.2024.106119.
[47] GUYON I, ELISSEEFF A. An Introduction to Variable and Feature Selection. Journal of Machine Learning Research, 2003, 3: 1157-1182.
[48] LI J D, CHENG K W, WANG S H, et al. Feature Selection: A Data Perspective. ACM Computing Surveys, 2017, 50(6). DOI: 10. 1145/31366.
[49] PENG H C, LONG F H, DING C. Feature Selection Based on Mutual Information Criteria of Max-Dependency, Max-Relevance, and Min-Redundancy. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27(8): 1226-1238.
[50] LEE S I, LEE H, ABBEEL P, et al. Efficient L1 Regularized Logistic Regression // Proc of the 21st National Conference on Artificial Intelligence. Palo Alto, USA: AAAI Press, 2006, I: 401-408.
[51] DEMARCHI L, KANIA A, CIEZKOWSKI W, et al. Recursive Feature Elimination and Random Forest Classification of Natura 2000 Grasslands in Lowland River Valleys of Poland Based on Airborne Hyperspectral and LiDAR Data Fusion. Remote Sensing, 2020, 12(11). DOI: 10.3390/rs12111842. |
|
|
|
2026, Vol. 39 
