Hybrid Deep Learning Strategies for Epileptic Seizure Detection

-, Mohammad Tabish (2024) Hybrid Deep Learning Strategies for Epileptic Seizure Detection. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (682kB) | Preview
Preview	PDF (Configuration Manual) Download (976kB) | Preview

Abstract

The health challenges related to epileptic seizures are of paramount concern and accurate identification at early stages is important in order to positively influence patient outcomes. Deep learning has revolutionized the detection, monitoring, and diagnosis of epileptic seizures to a greater extent in recent years, surging towards real-time processing. In this work, a novel deep learning method is proposed to detect epileptic seizures, which combines CNN (Convolutional Neural Networks) with LSTM or GRU. The research question is whether integrating spatial and temporal feature extraction via these hybrid models in an ensembled manner can improve the accuracy and dependability of seizure detection within EEG. The solution includes training these models on a set of EEG recordings showing healthy, interictal, and ictal states with significant pre-processing to normalize input signals. The CNN layers capture spatial features, while the LSTM and GRU layers handle temporal dependencies. Evaluation determined that the CNN-LSTM model produces superior accuracy compared with alternative configurations. A Flask web service is developed for real-time seizure detection, where users can upload EEG files to preprocess signals, predict seizures, and retrieve related information from Wikipedia. These findings confirm the efficiency of combining state-of-the-art deep learning models to enhance seizure detection, which will be advantageous in healthcare. Future work will focus on further refining the model's generalization abilities, considering multiple datasets, and investigating clinical deployment scenarios.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Tomer, Vikas UNSPECIFIED
Uncontrolled Keywords:	Epileptic Seizure Detection; Deep Learning; Hybrid Models; Convolutional Neural Networks; Neural Networks
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning R Medicine > RC Internal medicine > RC0321 Neuroscience. Biological psychiatry. Neuropsychiatry > Neurology. Diseases of the Nervous System.
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	06 Aug 2025 14:11
Last Modified:	06 Aug 2025 14:11
URI:	https://norma.ncirl.ie/id/eprint/8443

Actions (login required)

View Item