Applied Intelligence and Computing
Detecting the Seizure Conditions of Humans with EEG Dataset using Deep Belief Network Algorithm
Authors: Gunavathi M, Sudha S, Oviyaajanani S and Girija V
Publishing Date: 01-01-2025
ISBN: 978-81-955020-9-7
Abstract
Seizure detection is a critical aspect of epilepsy management, as timely intervention can significantly improve patient outcomes. This study presents a comprehensive investigation into the application of Deep Belief Network (DBN) and Recurrent Neural Network (RNN) algorithms for the automatic detection of seizure conditions in humans using EEG (Electroencephalogram) datasets. EEG data, known for its high temporal resolution, is particularly well-suited for capturing the dynamic patterns associated with seizures. In this research, we first describe the preprocessing steps applied to the EEG dataset, including noise removal, filtering, and feature extraction, aimed at enhancing the quality of the input data. To enhance the robustness of our approach, we explore ensemble techniques to combine the outputs of the two algorithms. Rigorous experiments are conducted, employing standard evaluation metrics such as sensitivity, specificity, and F1-score, with cross-validation to assess the model’s performance. Our results demonstrate the promise of the combined DBN and RNN approach in detecting seizure conditions with a high degree of accuracy. We provide comprehensive analyses of the experimental outcomes, including visualizations such as confusion matrices and ROC curves, and discuss the clinical implications of our findings. This research contributes to the growing body of knowledge in EEG-based seizure detection, offering insights into the potential for leveraging deep learning algorithms to improve the early detection and management of seizures in clinical settings
Keywords
Seizure detection; EEG signals; deep learning; feature selection.
Cite as
Gunavathi M, Sudha S, Oviyaajanani S and Girija V, "Detecting the Seizure Conditions of Humans with EEG Dataset using Deep Belief Network Algorithm", In: Mukesh Saraswat and Rajani Kumari (eds), Applied Intelligence and Computing, SCRS, India, 2025, pp. 237-246. https://doi.org/10.56155/978-81-955020-9-7-23