An Interpretable Machine Learning Framework for Interictal EEG-based Epilepsy Classification

Year 2026, Volume: 2 Issue: 1, 96 - 110, 30.01.2026

Reyhan Hoşavcı , Raziye Kübra Kumrular

https://doi.org/10.26650/d3ai.1854748

https://izlik.org/JA47WG25PC

Abstract

Reliable detection of interictal epileptic activity from electroencephalography (EEG) signals remains a challenging problem, as interictal patterns lack the distinctive characteristics observed during seizure episodes. Although many existing studies report high classification performance, this is often achieved by including ictal EEG signals, which substantially simplifies the task and limits clinical realism. In addition, the growing use of complex deep learning models has raised concerns regarding interpretability and practical deployment in clinical settings. To address these limitations, an interpretable machine learning framework is proposed for discriminating normal EEG signals from pathological interictal activity. EEG recordings from healthy subjects and epilepsy patients were analysed using a comprehensive set of features from the time and frequency domains designed to characterise spectral characteristics and temporal dynamics. Multiple machine learning classifiers were evaluated under a unified experimental protocol using five-fold cross-validation to ensure robust performance assessment. The experimental results demonstrate that the k-nearest neighbour classifier achieved the most balanced performance, with an accuracy of 95.5% in distinguishing interictal EEG segments from normal recordings. Furthermore, a Shapley-based feature contribution analysis revealed that a compact subset of influential features can retain high classification accuracy, indicating that model complexity can be significantly reduced without compromising performance. These findings demonstrate that effective interictal EEG discrimination can be achieved using a transparent and computationally efficient approach, providing an interpretable methodological benchmark for EEG-based analysis under controlled experimental conditions.

Keywords

Machine Learning , Explainable Artificial Intelligence , Electroencephalography (EEG) , Interictal Epilepsy Detecti

References

• Tracey A Milligan. 2021. Epilepsy: a clinical overview. The American Journal of Medicine 134, 7 (2021), 840–847. google scholar
• Valery L Feigin, Theo Vos, Balakrishnan Sukumaran Nair, Simon I Hay, Yohannes Habtegiorgis Abate, Abdallah H A Abd Al Magied, Samar Abd ElHafeez, Atef Abdelkader, Mohammad-Amin Abdollahifar, Auwal Abdullahi, Richard Gyan Aboagye, Lucas Guimarães Abreu, Samir Abu Rumeileh, Hasan Abualruz, Salahdein Aburuz, et al. 2025. Global, regional, and national burden of epilepsy, 1990–2021: a systematic analysis for the Global Burden of Disease Study 2021. The Lancet Public Health 10, 3 (March 2025), e203–e227. https://doi.org/10.1016/S2468-2667(24)00302-5 google scholar
• World Health Organization. 2022. Follow-up to the political declaration of the third high-level meeting of the General Assembly on the prevention and control of non-communicable diseases. World Health Organization, Geneva, Switzerland. google scholar
• Tao Zhang, Jichi Chen, and Kemal Polat. 2026. Enhanced epileptic seizure detection using CNNs with convolutional block attention and short-term memory networks. Behavioural Brain Research 496, (January 2026), 115831. https://doi.org/10.1016/j.bbr.2025.115831 google scholar
• Merete Osler, Erik L Mortensen, Kaare Christensen, and Gunhild T Christensen. 2018. A bidirectional association between cognitive ability in young adulthood and epilepsy: a population-based cohort study. International Journal of Epidemiology 47, 4 (August 2018), 1151–1158. https://doi.org/10.1093/ije/dyy018 google scholar
• Resmi Cherian and E. Gracemary Kanaga. 2022. Theoretical and methodological analysis of EEG based seizure detection and prediction: An exhaustive review. Journal of Neuroscience Methods 369, (March 2022), 109483. https://doi.org/10.1016/j.jneumeth.2022.109483 google scholar
• Lorenzo Frassineti, Angela Parente, and Claudia Manfredi. 2021. Multiparametric EEG analysis of brain network dynamics during neonatal seizures. Journal of neuroscience methods 348, (2021), 109003. google scholar
• Shivarudhrappa Raghu, Natarajan Sriraam, Yasin Temel, Shyam Vasudeva Rao, and Pieter L Kubben. 2020. EEG based multi-class seizure type classification using convolutional neural network and transfer learning. Neural Networks 124, (2020), 202–212. google scholar
• Emina Alickovic, Jasmin Kevric, and Abdulhamit Subasi. 2018. Performance evaluation of empirical mode decomposition, discrete wavelet transform, and wavelet packed decomposition for automated epileptic seizure detection and prediction. Biomedical signal processing and control 39, (2018), 94–102. google scholar
• Lijun Wei, Yuanyu Yu, Yuping Qin, and Shuang Zhang. 2025. A Survey of EEG-Based Approaches to Classroom Attention Assessment in Education. Information 16, 10 (2025). google scholar
• Md Khurram Monir Rabby, AKM Kamrul Islam, Saeid Belkasim, and Marwan U Bikdash. 2021. Wavelet transform-based feature extraction approach for epileptic seizure classification. In Proceedings of the 2021 ACM southeast conference, 2021. 164–169. google scholar
• Abdulhamit Subasi. 2007. EEG signal classification using wavelet feature extraction and a mixture of expert model. Expert Systems with Applications 32, 4 (2007), 1084–1093. google scholar
• Han-Tai Shiao, Vladimir Cherkassky, Jieun Lee, Brandon Veber, Edward E Patterson, Benjamin H Brinkmann, and Gregory A Worrell. 2016. SVM-based system for prediction of epileptic seizures from iEEG signal. IEEE Transactions on Biomedical Engineering 64, 5 (2016), 1011–1022. google scholar
• U Rajendra Acharya, Shu Lih Oh, Yuki Hagiwara, Jen Hong Tan, and Hojjat Adeli. 2018. Deep convolutional neural network for the automated detection and diagnosis of seizure using EEG signals. Computers in biology and medicine 100, (2018), 270–278. google scholar
• Yixuan Tang, Qianyi Wu, Haifeng Mao, and Lihua Guo. 2024. Epileptic seizure detection based on path signature and Bi-LSTM network with attention mechanism. IEEE Transactions on Neural Systems and Rehabilitation Engineering 32, (2024), 304–313. google scholar
• Meysam Golmohammadi, Saeedeh Ziyabari, Vinit Shah, Silvia Lopez de Diego, Iyad Obeid, and Joseph Picone. 2017. Deep architectures for automated seizure detection in scalp EEGs. arXiv preprint arXiv:1712.09776 (2017). google scholar
• Yang Li, Zuyi Yu, Yang Chen, Chunfeng Yang, Yue Li, X Allen Li, and Baosheng Li. 2020. Automatic seizure detection using fully convolutional nested LSTM. International journal of neural systems 30, 04 (2020), 2050019 google scholar
• Ralph G Andrzejak, Klaus Lehnertz, Florian Mormann, Christoph Rieke, Peter David, and Christian E Elger. 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 (2001), 061907. google scholar
• The MathWorks, Inc. 2021. MATLAB. The MathWorks, Inc., Natick, Massachusetts, United States. google scholar
• Thomas Cover and Peter Hart. 1967. Nearest neighbor pattern classification. IEEE transactions on information theory 13, 1 (1967), 21–27. google scholar
• Sunil Kumar Prabhakar, Young-Gi Ju, Harikumar Rajaguru, and Dong-Ok Won. 2022. Sparse measures with swarm-based pliable hidden Markov model and deep learning for EEG classification. Frontiers in Computational Neuroscience 16, (2022), 1016516. google scholar
• U Rajendra Acharya, Filippo Molinari, S Vinitha Sree, Subhagata Chattopadhyay, Kwan-Hoong Ng, and Jasjit S Suri. 2012. Automated diagnosis of epileptic EEG using entropies. Biomedical signal processing and control 7, 4 (2012), 401–408. google scholar
• Vladimir N Vapnik. 1999. An overview of statistical learning theory. IEEE transactions on neural networks 10, 5 (1999), 988–999. google scholar
• George H John and Pat Langley. 2013. Estimating continuous distributions in Bayesian classifiers. arXiv preprint arXiv:1302.4964 (2013). google scholar
• Ji Zhu and Trevor Hastie. 2005. Kernel logistic regression and the import vector machine. Journal of Computational and Graphical Statistics 14, 1 (2005), 185–205. google scholar
• Guanqing Kong, Shuang Ma, Wei Zhao, Haifeng Wang, Qingxi Fu, and Jiuru Wang. 2024. A novel method for optimizing epilepsy detection features through multi-domain feature fusion and selection. Frontiers in Computational Neuroscience 18, (2024), 1416838. google scholar
• Lloyd S Shapley and others. 1953. A value for n-person games. (1953). google scholar
• Scott M Lundberg and Su-In Lee. 2017. A unified approach to interpreting model predictions. Advances in neural information processing systems 30, (2017). google scholar
• Zeliha Kaya Akçelik, Reyhan Hoşavcı, Sümeyye Zülal Dik, Musa Aydin, and Zeki Kuş. 2025. Data Valuation with Shapley-based Methods for Medical Image Classification. In 2025 10th International Conference on Machine Learning Technologies (ICMLT), 2025. IEEE, 461–467.
• Christoph Molnar. 2020. Interpretable machine learning. Lulu.com. google scholar
• Umut Orhan, Mahmut Hekim, and Mahmut Ozer. 2011. EEG signals classification using the K-means clustering and a multilayer perceptron neural network model. Expert Systems with Applications 38, 10 (2011), 13475–13481. google scholar
• Mohammad Niknazar, SR Mousavi, B Vosoughi Vahdat, and Mohammad Sayyah. 2013. A new framework based on recurrence google scholar quantification analysis for epileptic seizure detection. IEEE journal of biomedical and health informatics 17, 3 (2013), 572–578. google scholar