Authors:
Sachin Chawla,Rajeev Ranjan,Yogendra Narayan,DOI NO:
https://doi.org/10.26782/jmcms.2026.05.00002Keywords:
Epileptic seizure detection,Electroencephalography (EEG) signals,Convolutional Neural Network,Vision Transformer,Adaptive Multi-Stream Feature Fusion,Abstract
Objective: Automated seizure detection from scalp electroencephalography(EEG) remains challenging because EEG signals are non-stationary, noisy, highly imbalanced, and vary substantially across patients. This study aimed to develop a robust deep learning framework for seizure detection under clinically relevant, leakage-controlled evaluation settings. Methods: We proposed a frequency-aware CNN–Vision Transformer (FA-CNNViT) framework integrating deterministic dataset harmonization, subject-wise leakage-controlled cross-validation, split-specific preprocessing, and post-split window generation. The model combines convolutional encoding for local morphological features with transformer-based modeling of long-range dependencies. An adaptive multi-stream feature fusion module was used to preserve temporal, spectral, and spatial information. Asymmetric focal loss addressed class imbalance, and Monte Carlo dropout was used to estimate predictive uncertainty. Results: On the CHB-MIT dataset, FA-CNNViT achieved 99.13% accuracy, 99.10% sensitivity, 99.47% specificity, 99.55% F1-score, and 99.74% ROC-AUC. In the cross-subject setting on the Turkish EEG dataset, it achieved 89.98% accuracy, 87.41% sensitivity, 88.61% specificity, 87.44% F1- score, and 88.79% ROC-AUC. Conclusion: The proposed framework achieved strong subject-wise performance and competitive cross-subject performance under a leakage-controlled evaluation protocol. Further refinement of false-positive control and prospective validation is needed before real-time clinical deployment.Refference:
I. Chen, Wenna, et al. "An automated detection of epileptic seizures EEG using CNN classifier based on feature fusion with high accuracy." BMC Medical informatics and Decision making 23.1 (2023): 96. 10.1186/s12911-023-02180-w
II. Christou, Vasileios, et al. "Evaluating the window size’s role in automatic EEG epilepsy detection." Sensors 22.23 (2022): 9233. 10.3390/s22239233
III. Ein Shoka, Athar A., et al. "EEG seizure detection: concepts, techniques, challenges, and future trends." Multimedia tools and applications 82.27 (2023): 42021-42051. 10.1007/s11042-023-15052-2
IV. Feng, Lufeng, et al. "A multi-view neural framework with attention for epileptic seizure classification." Journal of Neural Engineering 23.1 (2026): 016018. 10.1088/1741-2552/ae33f8
V. Guhdar, Mohammed, Ramadhan J. Mstafa, and Abdulhakeem O. Mohammed. "A multimodal temporal attention network for seizure classification via one-dimensional convolutional neural architecture." Biomedical Signal Processing and Control 112 (2026): 108495. 10.1016/j.bspc.2025.108495
VI. Islam, Md Rabiul, et al. "Epileptic seizure focus detection from interictal electroencephalogram: a survey." Cognitive neurodynamics 17.1 (2023): 1-23. 10.1007/s11571-022-09816-z
VII. Kashefi Amiri, Homa, Masoud Zarei, and Mohammad Reza Daliri. "Epileptic seizure detection from electroencephalogram signals based on 1D CNN-LSTM deep learning model using discrete wavelet transform." Scientific Reports 15.1 (2025): 32820. 10.1038/s41598-025-18479-9
VIII. Kode, Hepseeba, Khaled Elleithy, and Laiali Almazaydeh. "Epileptic seizure detection in EEG signals using machine learning and deep learning techniques." IEEE access 12 (2024): 80657-80668. 10.1109/ACCESS.2024.3409581
IX. Lebal, Abdelhamid, Abdelouahab Moussaoui, and Abdelmounaam Rezgui. "Epilepsy-Net: attention-based 1D-inception network model for epilepsy detection using one-channel and multi-channel EEG signals." Multimedia tools and applications 82.11 (2023): 17391-17413. 10.1007/s11042-022-13947-0
X. Li, Qi, Wei Cao, and Anyuan Zhang. "Multi-stream feature fusion of vision transformer and CNN for precise epileptic seizure detection from EEG signals." Journal of Translational Medicine 23.1 (2025): 871. 10.1186/s12967-025-06862-z
XI. Li, Yang, et al. "Automatic seizure detection using fully convolutional nested LSTM." International journal of neural systems 30.04 (2020): 2050019. 10.1142/S0129065720500197
XII. Luo, Weitao, et al. "EEG-Based Brain-Computer Interface: Fundamentals, Methods, Applications, and Challenges." IEEE Internet of Things Journal (2025). 10.1109/JIOT.2025.3625060
XIII. Ma, Mengnan, et al. "Research on epileptic EEG recognition based on improved residual networks of 1-D CNN and indRNN." BMC Medical Informatics and Decision Making 21.Suppl 2 (2021): 100. 10.1186/s12911-021-01438-5
XIV. Pan, Yayan, et al. "Downsampling of EEG signals for deep learning-based epilepsy detection." IEEE Sensors Letters 7.12 (2023): 1-4. 10.1109/LSENS.2023.3332392
XV. Ren, Juntao, et al. "An Event-Based Filtering and Weighted Enhanced Deep Learning Epileptic Seizure Prediction Method." Neural Networks (2025): 108424. 10.1016/j.neunet.2025.108424
XVI. Salini, G. Indu, I. Sowmy, and T. K. Sreeja. "FrAdadelta-CSA: Fractional Adadelta Chameleon Swarm Algorithm-based feature selection with SpikeGoogle-DenseNet for epileptic seizure detection." Computational Biology and Chemistry 119 (2025): 108550. 10.1016/j.compbiolchem.2025.108550
XVII. Shah, Syed Yaseen, et al. "Epileptic seizure classification based on random neural networks using discrete wavelet transform for electroencephalogram signal decomposition." Applied Sciences 14.2 (2024): 599. 10.3390/app14020599
XVIII. Shanmugam, Shalini, and Selvathi Dharmar. "A CNN-LSTM hybrid network for automatic seizure detection in EEG signals." Neural Computing and Applications 35.28 (2023): 20605-20617. 10.1007/s00521-023-08832-2
XIX. Srinivasan, Saravanan, et al. "Detection and classification of adult epilepsy using hybrid deep learning approach." Scientific reports 13.1 (2023): 17574. 10.1038/s41598-023-44763-7
XX. Sunaryono, Dwi, Riyanarto Sarno, and Joko Siswantoro. "Gradient boosting machines fusion for automatic epilepsy detection from EEG signals based on wavelet features." Journal of King Saud University-Computer and Information Sciences 34.10 (2022): 9591-9607. 10.1016/j.jksuci.2021.11.015
XXI. Tasci, Irem, et al. "Epilepsy detection in 121 patient populations using hypercube pattern from EEG signals." Information Fusion 96 (2023): 252-268. 10.1016/j.inffus.2023.03.022
XXII. Tawhid, Md Nurul Ahad, Siuly Siuly, and Tianning Li. "A convolutional long short-term memory-based neural network for epilepsy detection from EEG." IEEE Transactions on Instrumentation and Measurement 71 (2022): 1-11. 10.1109/TIM.2022.3217515
XXIII. Wang, Quanhong, et al. "A hybrid SVM and kernel function-based sparse representation classification for automated epilepsy detection in EEG signals." Neurocomputing 562 (2023): 126874. 10.1016/j.neucom.2023.126874
XXIV. Wang, Zhuohan, et al. "EEG-based seizure detection using dual-branch CNN-ViT network integrating phase and power spectrograms." Brain sciences 15.5 (2025): 509. 10.3390/brainsci15050509
XXV. Xu, Gaowei, et al. "A one-dimensional CNN-LSTM model for epileptic seizure recognition using EEG signal analysis." Frontiers in neuroscience 14 (2020): 578126. 10.3389/fnins.2020.578126