Comparative Analysis of 1D-CNN and BiLSTM Models in EEG-Based Epilepsy Detection

Öz

This study was developed to automatically detect epilepsy-specific signal features from electroencephalography (EEG) signals. The widely used Bonn EEG dataset was employed, reorganized for binary classification with normal and epileptic samples. In preprocessing, band-pass and notch filters were applied, the sampling rate was adjusted, windows were generated from the recordings, and each window was z-score normalized. In this study, two deep learning approaches were trained and compared under the same experimental setup: a 1D Convolutional Neural Network (1D-CNN) that captures local time-series patterns, and a Bidirectional Long Short-Term Memory (BiLSTM) that models temporal dependencies. Model performance was evaluated using metrics such as accuracy, precision, recall, F1-score, and ROC-AUC at both the window level and the segment level obtained via the mean probabilities of the windows. To statistically validate the reliability of the results, the models were tested using the 5-fold cross-validation method. Experimental findings indicate that the 1D-CNN model provides significantly higher accuracy rates compared to the BiLSTM model and exhibits a much more stable performance based on variance analyses.

Anahtar Kelimeler

• EGG
• Epilepsy
• Bonn Dataset
• 1D-CNN
• Bi-LSTM

EEG Tabanlı Epilepsi Tespitinde 1D-CNN ve BiLSTM Modellerinin Karşılaştırmalı Analizi

Öz

Bu çalışma, elektroensefalografi (EEG) sinyalleri üzerinden epilepsiye özgü sinyal özelliklerinin otomatik olarak tespit edilmesi amacıyla geliştirilmiştir. Kullanılan veri seti, literatürde yaygın biçimde başvurulan Bonn EEG veri seti olup normal ve epileptik örnekler ikili sınıflama çerçevesinde yeniden düzenlenmiştir. Önişlemede bant geçiren ve çentik (notch) filtre uygulanmış, örnekleme frekansı düzenlenmiş, kayıtlardan pencereler oluşturulmuş ve her pencere z-skor yöntemiyle normalize edilmiştir. Bu çalışmada iki derin öğrenme yaklaşımı aynı deney düzeni altında eğitilip karşılaştırılmıştır: yerel zaman serisi örüntülerini yakalayan 1B-Evrişimsel Sinir Ağı (1D-CNN) ve zamansal bağımlılıkları modelleyen çift yönlü Uzun Kısa Süreli Bellek (BiLSTM). Modellerin başarımı doğruluk, kesinlik, duyarlılık, F1 skoru ve ROC-AUC gibi ölçütlerle hem pencere düzeyinde hem de pencerelerin ortalama olasılıklarıyla elde edilen segment düzeyinde değerlendirilmiştir. Sonuçların güvenilirliğini istatistiksel olarak doğrulamak amacıyla modeller 5-katlı çapraz doğrulama (5-Fold Cross-Validation) yöntemiyle test edilmiştir. Deneysel bulgular, 1D-CNN modelinin BiLSTM modeline kıyasla belirgin şekilde daha yüksek doğruluk oranı sağladığını ve varyans analizlerine göre çok daha kararlı (stabil) bir performans sergilediğini ortaya koymaktadır.

Anahtar Kelimeler

• EGG
• Epilepsi
• Bonn Veri Seti
• 1D-CNN
• Bi-LSTM

Kaynakça

1. Abdullayeva E., Kahramanlı Örnek H. Diagnosing epilepsy from EEG using machine learning and Welch spectral analysis. Traitement du Signal 2024; 41(2): 971–977.
2. Andrzejak RG., Lehnertz K., Mormann F., Rieke C., David P., Elger CE. 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 2001; 64(6): 061907.
3. Bilen B. Karışıklık matrisi (Confusion Matrix). Medium t.y. [Erişim adresi: https://burhanbilen.medium.com/kar%C4%B1%C5%9F%C4%B1kl%C4%B1k-matrisi-confusion-matrix-990dfc718653] (Erişim Tarihi: 14.12.2025).
4. Cao X., Zheng S., Zhang J., Chen W., Du G. A hybrid CNN-Bi-LSTM model with feature fusion for accurate epilepsy seizure detection. BMC Medical Informatics and Decision Making 2025; 25(1): 6.
5. Chen W., Wang Y., Ren Y., Jiang H., Du G., Zhang J., Li J. An automated detection of epileptic seizures EEG using CNN classifier based on feature fusion with high accuracy. BMC Medical Informatics and Decision Making 2023; 23(1): 96.
6. Esha OK., Begum N., Rahman S. EpiNet: A hybrid machine learning model for epileptic seizure prediction using EEG signals from a 500 patient dataset. International Journal of Advanced Computer Science and Applications 2024; 15(1): 1182–1191.
7. Fisher RS., Acevedo C., Arzimanoglou A., Bogacz A., Cross JH., Elger CE., Engel J. Jr., Forsgren L., French JA., Glynn M., Hesdorffer DC., Lee BI., Mathern GW., Moshé SL., Perucca E., Scheffer IE., Tomson T., Watanabe M., Wiebe S. ILAE official report: A practical clinical definition of epilepsy. Epilepsia 2014; 55(4): 475–482.
8. Hazarika U., Borah BB., Roy S., Saikia MJ. Automated detection of aberrant episodes in epileptic conditions: Leveraging EEG and machine learning algorithms. Bioengineering 2025; 12(4): 355.

9. Huang L., Zhou K., Chen S., Chen Y., Zhang J. Automatic detection of epilepsy from EEGs using a temporal convolutional network with a self-attention layer. BioMedical Engineering OnLine 2024; 23(1): 50.
10. Khan SU., Jan SU., Koo I. Robust epileptic seizure detection using long short-term memory and feature fusion of compressed time–frequency EEG images. Sensors 2023; 23(23): 9572.
11. LeCun Y., Bengio Y., Hinton G. Deep learning. Nature 2015; 521(7553): 436–444.
12. Mirowski P., Madhavan D., LeCun Y., Kuzniecky R. Classification of patterns of EEG synchronization for seizure prediction. Clinical Neurophysiology 2009; 120(11): 1927–1940.
13. Örgerim A., Tunç Abubakar T., Tokmak M. Performance comparison of neural networks: A case of data scientists’ job change prediction. Osmaniye Korkut Ata University Journal of The Institute of Science and Technology 2025; 8(3): 1100–1119.
14. Rashed-Al-Mahfuz M., Moni MA., Uddin S., Alyami SA., Summers MA., Eapen V. A deep convolutional neural network method to detect seizures and characteristic frequencies using epileptic electroencephalogram (EEG) data. IEEE Journal of Translational Engineering in Health and Medicine 2021; 9(1): 2000112.
15. Srinivasan S., Dayalane S., Mathivanan SK., Rajadurai H., Jayagopal P., Dalu GT. Detection and classification of adult epilepsy using hybrid deep learning approach. Scientific Reports 2023; 13(1): 17574.
16. Tran LV., Tran HM., Le TM., Huynh TTM., Tran HT., Dao SVT. Application of machine learning in epileptic seizure detection. Diagnostics 2022; 12(11): 2879.