Epileptic Seizures Detection Using Continuous Time Wavelet Based Artificial Neural Networks

Abstract

The aim of this work is to develop a new method for automatic detection and classification of EEG patterns using continuous wavelet transforms (CWT) and artificial neural networks (ANN). Our method consists of EEG data selection, feature extraction and classification stage. For the data selection we use temporal lobe seizures for evaluation recorded from patients during 84 hours at hospital. In feature extraction stage we use best basis mother wavelet functions and wavelet thresholding technique. In classification stage we implement multi layer perceptron neural networks according to standard backpropogation algorithm. We demonstrate the efficiency of our wavelet based feature extraction method on data to improve the ANN classification performance. We achieved 95.8% accuracy in classification of ictal and interictal EEG segments.

More by Berdakh Abibullaev

• Onset Review: traceable analysis of high-frequency oscillations and seizure onset in intracranial EEGZenodo · 2026
• Multi-scale EEG feature decoding with Swin Transformers for subject independent motor imagery BCIsScientific Reports · 2026 · Citations: 12
• Generative data augmentation for machine learning-based quantitative cancer cell detection using optical fiber biosensorsEPJ Web of Conferences · 2026
• Deep learning in intracranial EEG for seizure detection: advances, challenges, and clinical applicationsFrontiers in Neuroscience · 2025 · Citations: 13