A Feature Selection and Hyperparameter Optimization Framework for Multiclass EEG-based Epilepsy Detection

Seda Şaşmaz Karacan

doi:10.35377/saucis...1832685

A Feature Selection and Hyperparameter Optimization Framework for Multiclass EEG-based Epilepsy Detection

Abstract

Epilepsy is a neurological disorder characterized by recurrent seizures, and EEG continues to be the gold standard for diagnosis in clinical practice. However, the nonlinear nature of the EEG and its sensitivity to noise make manual interpretations difficult. This study presents a feature selection and hyperparameter optimization approach for automatic multi-class EEG-based epilepsy detection. Five classes representing two healthy states and three epileptic states were analyzed using EEG signals from the Bonn database. Temporal, statistical, fractal, and spectral features were extracted, and their importance was ranked according to ANOVA F-values. Classifications were performed using support vector machines, random forests, nearest-neighbor algorithms, decision trees, multilayer perceptrons, and linear discriminant analysis; hyperparameters were optimized using Grid Search Cross-Validation. The robustness of the proposed model was ensured using grouped layered k-fold cross-validation. With this approach, it achieved an F1-score of 0.8126 in 5-class classification and 0.9894 in binary classification, outperforming existing methods on the same dataset. Hjorth complexity, standard deviation, and Hjorth mobility emerged as the most discriminative features for epileptic states. The study demonstrates that combining feature selection with hyperparameter optimization provides a robust, interpretable framework for multi-class EEG classification. This approach has the potential to support neurologists in their epilepsy assessments by improving diagnostic accuracy and providing clinically meaningful insights.

Keywords

Ethical Statement

It is declared that during the preparation of this study, scientific and ethical principles were followed, and all studies used are listed in the bibliography.

References

T. A. Milligan, “Epilepsy: A clinical overview,” Amer. J. Med., vol. 134, no. 7, pp. 840–847, 2021, doi: 10.1016/j.amjmed.2021.01.038. [Online]. Available: https://doi.org/10.1016/j.amjmed.2021.01.038
World Health Organization, “Epilepsy,” 2025. [Online]. Available: https://www.who.int/news-room/fact-sheets/detail/epilepsy. [Accessed: Aug. 14, 2025].
T. Shan et al., “Global, regional, and national time trends in the burden of epilepsy, 1990–2019: An age-period-cohort analysis for the global burden of disease 2019 study,” Front. Neurol., vol. 15, Art. no. 1418926, 2024, doi: 10.3389/fneur.2024.1418926. [Online]. Available: https://doi.org/10.3389/fneur.2024.1418926
J. Moriarty, “Anxiety disorders and epilepsy,” in Introduction to Epilepsy, G. Alarcón and A. Valentín, Eds. Cambridge, U.K.: Cambridge Univ. Press, 2012, pp. 518–519.
M. K. Siddiqui, R. Morales-Menendez, X. Huang, and N. Hussain, “A review of epileptic seizure detection using machine learning classifiers,” Brain Informat., vol. 7, Art. no. 5, 2020, doi: 10.1186/s40708-020-00105-1. [Online]. Available: https://doi.org/10.1186/s40708-020-00105-1
P. B. Boz and C. Özden, “Decision support system based on ensemble models in distinguishing epilepsy types,” Epilepsy Behav., vol. 170, Art. no. 110470, 2025, doi: 10.1016/j.yebeh.2025.110470. [Online]. Available: https://doi.org/10.1016/j.yebeh.2025.110470
H. Bao and J. Ma, “A novel frame funnel-shaped CNN with SE block for epilepsy detection in EEG signals,” in Proc. 10th Int. Conf. Intell. Comput. Signal Process. (ICSP), Xi’an, China, 2025, pp. 471–477, doi: 10.1109/ICSP65755.2025.11087152. [Online]. Available: https://doi.org/10.1109/ICSP65755.2025.11087152
S. S. Baawi, E. Hakem, A. A. Al-Hamzawi, D. Al-Shammary, A. Ibaida, and A. M. Mahdi, “Hilbert similarity convex for efficient EEG feature selections,” J. Supercomput., vol. 81, 2025, doi: 10.1007/s11227-025-07199-6. [Online]. Available: https://doi.org/10.1007/s11227-025-07199-6

M. M. Qureshi and M. Kaleem, “EEG-based seizure prediction with machine learning,” Signal Image Video Process., vol. 17, pp. 1543–1554, 2023, doi: 10.1007/s11760-022-02363-4. [Online]. Available: https://doi.org/10.1007/s11760-022-02363-4
C. Shi, J. Gao, J. Yu, L. Zhao, and F. Jia, “A novel similarity-constrained feature selection method for epilepsy detection via EEG signals,” J. King Saud Univ.—Comput. Inf. Sci., vol. 37, Art. no. 141, 2025, doi: 10.1007/s44443-025-00152-w. [Online]. Available: https://doi.org/10.1007/s44443-025-00152-w
Z. Huang, L. Zhang, J. Xu, Y. Wu, M. Chen, and M. Zhao, “EEG detection and recognition model for epilepsy based on dual attention mechanism,” Sci. Rep., vol. 15, Art. no. 9404, 2025, doi: 10.1038/s41598-025-90315-6. [Online]. Available: https://doi.org/10.1038/s41598-025-90315-6
J. Su, Z. Huang, Y. Ma, H. Shi, Y. Yang, M. Xi, and B. Li, “A model for epileptic EEG detection and recognition based on multi-attention mechanism and spatiotemporal features,” Sci. Rep., vol. 15, Art. no. 31993, 2025, doi: 10.1038/s41598-025-17256-y. [Online]. Available: https://doi.org/10.1038/s41598-025-17256-y
Y. Berrich and Z. Guennoun, “EEG-based epilepsy detection using CNN-SVM and DNN-SVM with feature dimensionality reduction by PCA,” Sci. Rep., vol. 15, Art. no. 14313, 2025, doi: 10.1038/s41598-025-95831-z. [Online]. Available: https://doi.org/10.1038/s41598-025-95831-z
M. Sadiq, M. N. Kadhim, D. Al-Shammary, and M. Milanova, “Novel EEG feature selection based on Hellinger distance for epileptic seizure detection,” Smart Health, vol. 35, Art. no. 100536, 2025, doi: 10.1016/j.smhl.2024.100536. [Online]. Available: https://doi.org/10.1016/j.smhl.2024.100536
S. L. Alzamili, S. S. Baawi, M. N. Kadhim, D. Al-Shammary, and A. Ibaida, “Efficient feature selection based on Ruzicka similarity for EEG diagnosis,” Int. J. Inf. Technol., vol. 17, pp. 3373–3387, 2025, doi: 10.1007/s41870-025-02572-3. [Online]. Available: https://doi.org/10.1007/s41870-025-02572-3
S. M. Malakouti, “Enhanced epilepsy detection using discrete wavelet transform and bandpass filtering on EEG data: Integration of ART-based and LVQ models,” Clin. eHealth, vol. 8, pp. 134–145, 2025, doi: 10.1016/j.ceh.2025.07.001. [Online]. Available: https://doi.org/10.1016/j.ceh.2025.07.001
R. G. Andrzejak, K. Lehnertz, F. Mormann, C. Rieke, P. David, and C. E. Elger, “Indications of nonlinear deterministic and finite-dimensional structures in time series of brain electrical activity,” Phys. Rev. E, vol. 64, no. 6, Art. no. 061907, 2001, doi: 10.1103/PhysRevE.64.061907. [Online]. Available: https://doi.org/10.1103/PhysRevE.64.061907
P. Virtanen et al., “SciPy 1.0: Fundamental algorithms for scientific computing in Python,” Nature Methods, vol. 17, pp. 261–272, 2020, doi: 10.1038/s41592-019-0686-2. [Online]. Available: https://doi.org/10.1038/s41592-019-0686-2
N. Avital, E. Nahum, G. C. Levi, and D. Malka, “Cognitive state classification using convolutional neural networks on gamma-band EEG signals,” Appl. Sci., vol. 14, no. 18, Art. no. 8380, 2024, doi: 10.3390/app14188380. [Online]. Available: https://doi.org/10.3390/app14188380
E. M. Whitham et al., “Scalp electrical recording during paralysis: Quantitative evidence that EEG frequencies above 20 Hz are contaminated by EMG,” Clin. Neurophysiol., vol. 118, no. 8, pp. 1877–1888, 2007, doi: 10.1016/j.clinph.2007.04.027. [Online]. Available: https://doi.org/10.1016/j.clinph.2007.04.027
S. D. Muthukumaraswamy, “High-frequency brain activity and muscle artifacts in MEG/EEG: A review and recommendations,” Front. Hum. Neurosci., vol. 7, Art. no. 138, 2013, doi: 10.3389/fnhum.2013.00138. [Online]. Available: https://doi.org/10.3389/fnhum.2013.00138
H. Göker, “Comparison of different spectral analysis methods with an experimental EEG dataset,” AINTELIA Sci. Notes J., vol. 1, no. 1, 2022, doi: 10.5281/zenodo.8070931. [Online]. Available: https://doi.org/10.5281/zenodo.8070931
L. A. Moctezuma and M. Molinas, “Classification of low-density EEG for epileptic seizures by energy and fractal features based on EMD,” J. Biomed. Res., vol. 34, no. 3, pp. 180–190, 2020, doi: 10.7555/JBR.33.20190009. [Online]. Available: https://doi.org/10.7555/JBR.33.20190009
I. Wijayanto, R. Hartanto, and H. A. Nugroho, “Comparison of empirical mode decomposition and coarse-grained procedure for detecting pre-ictal and ictal condition in electroencephalography signal,” Informat. Med. Unlocked, vol. 19, Art. no. 100325, 2020, doi: 10.1016/j.imu.2020.100325. [Online]. Available: https://doi.org/10.1016/j.imu.2020.100325
G. Kaushik, P. Gaur, R. R. Sharma, and R. B. Pachori, “EEG signal based seizure detection focused on Hjorth parameters from tunable-Q wavelet sub-bands,” Biomed. Signal Process. Control, vol. 76, Art. no. 103645, 2022, doi: 10.1016/j.bspc.2022.103645. [Online]. Available: https://doi.org/10.1016/j.bspc.2022.103645
T. Cecchin, R. Ranta, L. Koessler, O. Caspary, H. Vespignani, and L. Maillard, “Seizure lateralization in scalp EEG using Hjorth parameters,” Clin. Neurophysiol., vol. 121, no. 3, pp. 290–300, 2010, doi: 10.1016/j.clinph.2009.10.033. [Online]. Available: https://doi.org/10.1016/j.clinph.2009.10.033
A. M. Murro et al., “Localization of temporal lobe seizures with quantitative EEG,” Electroencephalogr. Clin. Neurophysiol., vol. 86, no. 2, pp. 88–93, 1993, doi: 10.1016/0013-4694(93)90080-F. [Online]. Available: https://doi.org/10.1016/0013-4694(93)90080-F
M. M. Rashid and M. Ahmad, “Epileptic seizure classification using statistical features of EEG signal,” in Proc. Int. Conf. Elect., Comput. Commun. Eng. (ECCE), Cox’s Bazar, Bangladesh, 2017, pp. 308–312, doi: 10.1109/ECACE.2017.7912923. [Online]. Available: https://doi.org/10.1109/ECACE.2017.7912923
G. Yogarajan, P. Gupta, R. Pal, and S. Rathore, “EEG-based epileptic seizure detection using binary dragonfly algorithm and deep neural network,” Sci. Rep., vol. 13, Art. no. 17710, 2023, doi: 10.1038/s41598-023-44318-w. [Online]. Available: https://doi.org/10.1038/s41598-023-44318-w
J. Xiang, X. Zhang, W. Li, L. Chen, and Y. Wang, “Kurtosis and skewness of high-frequency brain signals are altered in pediatric epilepsy,” Brain Commun., vol. 2, no. 1, 2020, doi: 10.1093/braincomms/fcaa036. [Online]. Available: https://doi.org/10.1093/braincomms/fcaa036
O. K. Fasil and R. Rajesh, “Time-domain exponential energy for epileptic EEG signal classification,” Neurosci. Lett., vol. 694, pp. 1–8, 2019, doi: 10.1016/j.neulet.2018.10.062. [Online]. Available: https://doi.org/10.1016/j.neulet.2018.10.062
F. Pedregosa et al., “Scikit-learn: Machine learning in Python,” J. Mach. Learn. Res., vol. 12, pp. 2825–2830, 2011.
M. M. Ali, M. S. Islam, M. N. Uddin, and M. A. Uddin, “A conceptual IoT framework based on Anova-F feature selection for chronic kidney disease detection using deep learning approach,” Intell.-Based Med., vol. 10, Art. no. 100170, 2024, doi: 10.1016/j.ibmed.2024.100170. [Online]. Available: https://doi.org/10.1016/j.ibmed.2024.100170
J. Matos, G. Peralta, J. Heyse, E. Menetre, M. Seeck, and P. van Mierlo, “Diagnosis of epilepsy with functional connectivity in EEG after a suspected first seizure,” Bioengineering, vol. 9, no. 11, Art. no. 690, 2022, doi: 10.3390/bioengineering9110690. [Online]. Available: https://doi.org/10.3390/bioengineering9110690

Details

Primary Language

English

Subjects

Artificial Intelligence (Other)

Journal Section

Research Article

Authors

Seda Şaşmaz Karacan ^*
0000-0002-0334-260X
Türkiye

Publication Date

September 30, 2026

Submission Date

November 29, 2025

Acceptance Date

March 13, 2026

Published in Issue

Year 2026 Volume: 9 Number: 4

DOI

https://doi.org/10.35377/saucis...1832685

IZ

https://izlik.org/JA26ZM97XL

Cite

RIS / Bibtex

APA

Şaşmaz Karacan, S. (2026). A Feature Selection and Hyperparameter Optimization Framework for Multiclass EEG-based Epilepsy Detection. Sakarya University Journal of Computer and Information Sciences, 9(4), 1294-1304. https://doi.org/10.35377/saucis...1832685

AMA

1.Şaşmaz Karacan S. A Feature Selection and Hyperparameter Optimization Framework for Multiclass EEG-based Epilepsy Detection. SAUCIS. 2026;9(4):1294-1304. doi:10.35377/saucis.1832685

Chicago

Şaşmaz Karacan, Seda. 2026. “A Feature Selection and Hyperparameter Optimization Framework for Multiclass EEG-Based Epilepsy Detection”. Sakarya University Journal of Computer and Information Sciences 9 (4): 1294-1304. https://doi.org/10.35377/saucis. 1832685.

EndNote

Şaşmaz Karacan S (September 1, 2026) A Feature Selection and Hyperparameter Optimization Framework for Multiclass EEG-based Epilepsy Detection. Sakarya University Journal of Computer and Information Sciences 9 4 1294–1304.

IEEE

[1]S. Şaşmaz Karacan, “A Feature Selection and Hyperparameter Optimization Framework for Multiclass EEG-based Epilepsy Detection”, SAUCIS, vol. 9, no. 4, pp. 1294–1304, Sept. 2026, doi: 10.35377/saucis...1832685.

ISNAD

Şaşmaz Karacan, Seda. “A Feature Selection and Hyperparameter Optimization Framework for Multiclass EEG-Based Epilepsy Detection”. Sakarya University Journal of Computer and Information Sciences 9/4 (September 1, 2026): 1294-1304. https://doi.org/10.35377/saucis. 1832685.

JAMA

1.Şaşmaz Karacan S. A Feature Selection and Hyperparameter Optimization Framework for Multiclass EEG-based Epilepsy Detection. SAUCIS. 2026;9:1294–1304.

MLA

Şaşmaz Karacan, Seda. “A Feature Selection and Hyperparameter Optimization Framework for Multiclass EEG-Based Epilepsy Detection”. Sakarya University Journal of Computer and Information Sciences, vol. 9, no. 4, Sept. 2026, pp. 1294-0, doi:10.35377/saucis. 1832685.

Vancouver

1.Seda Şaşmaz Karacan. A Feature Selection and Hyperparameter Optimization Framework for Multiclass EEG-based Epilepsy Detection. SAUCIS. 2026 Sep. 1;9(4):1294-30. doi:10.35377/saucis. 1832685