An epilepsy prediction and management system based on federated learning combined with hybrid harmony search and mutual information (HSA-MI)-based feature selection approach

Sep 30, 2026

Uploaded by Melati

Epilepsy is a widespread neurological disorder marked by unpredictable seizures that disrupt daily life and require accurate, timely detection to support intervention and quality of life. Existing EEG seizure detection approaches often suffer from limited generalization, excessive incorrect alerts, and inefficient feature-selection pipelines, while conventional deep learning may underperform on EEG patterns with both temporal and spatial characteristics. This study presents EpilepNet-LD with federated learning for privacy-preserving decentralized EEG analysis, using HSA-MI feature selection to optimize temporal, spectral, and spatial features and reduce computational overhead.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages26

Reading Time1 h 6 min

AccessFree to download

Introduction

Proposed approach: EpilepNet-LD with federated learning

HSA-MI feature selection

Hybrid LSTM and DenseNet-121 architecture

Experimental evaluation and reported performance

System goals and clinical management module

[www. nature.com/scientificreports](www. nature.com/scientificreports) OPEN An epilepsy prediction and management system based on federated learning combined with hybrid harmony search and mutual information (HSA-MI)-based feature selection approach Mohd Abdul Rahim Khan1􀀍, Khaled Mahmoud Heba2,5, Abbass Hassan Abbass3,5, Mujeeb Ahmed Shaikh4,5, GhanshyamG. Tejani6,7􀀍, Sunil Kumar Sharma8 & Mohammad Khaleel Sallam Ma’aitah9 Epilepsy represents a widespread neurological disorder that causes unexpected seizure occurrences which produce significant obstacles for daily life activities in people worldwide. Real-time seizure detection accuracy stands vital for safeguarding patients while ensuring timely interventions and bettering their quality of life. The present seizure detection methods suffer from multiple limitations including poor applicability across different cases and excessive incorrect alerts and inefficient featureselection procedures. Traditional deep learning algorithms generally fail to detect EEG patterns with both temporal and spatial characteristics thereby producing subpar results in practical settings. This study proposes EpilepNet-LD integrated with federated learning (FL), a novel framework for decentralized EEG analysis that preserves patient privacy while enabling collaborative model training. The framework employs a hybrid harmony search and mutual information (HSA-MI) feature selection technique to identify optimal temporal, spectral, and spatial EEG features, reducing computational overhead. The EpilepNet-LD architecture combines long short-term memory (LSTM) networks to capture temporal dependencies with DenseNet-121 to extract hierarchical spatial features, improving seizure detection performance. Extensive experimental evaluations confirmed the proposed method achieves 99.41% extensive experiments demonstrate that the proposed method achieves 99.41% accuracy and 99.50% sensitivity, outperforming existing state-of-the-art approaches. The integrated FL and HSA-MI approach, coupled with the EpilepNet-LD classifier, enables robust, real-time, and high-precision seizure detection, offering a reliable solution for advanced epilepsy monitoring and management. Keywords Epilepsy prediction, Federated learning, Hybrid harmony search and mutual information, EpilepNet-LD 1Department of Electrical Engineering and Computer Science, College of Engineering, A’sharqiyah University, Ibra 400, Sultanate of Oman. 2Department of Architecture, College of Engineering and Information Technology, Onaiza Colleges, Unaizah, Saudi Arabia. 3Department of Electrical Engineering, College of Engineering and Information Technology, Onaiza Colleges, Unaizah, Saudi Arabia. 4Department of Basic Medical Science, College of Medicine, AlMaarefa University, Ad Diriyah, Riyadh 13713, Saudi Arabia. 5King Salman Center for Disability Research, Riyadh 11614, Saudi Arabia. 6Department of Research Analytics, Saveetha Dental College and Hospitals, Saveetha Institute of Medical and Technical Sciences, Saveetha University, Chennai 600077, India. 7Department of Industrial Engineering and Management,Yuan Ze University, Taoyuan 320315, Taiwan. 8Department of Information Systems/ College of Computer and Information Sciences, Majmaah University, Majmaah 11952, Saudi Arabia. 9Faculty of Engineering and Technology, Department of Robotics and Artificial Intelligence, Applied Science Private University, Amman 11937, Jordan. 􀀍 [email: mohd.khan@asu.edu.om](email: mohd.khan@asu.edu.om); [p.shyam23@gmail.com](p.shyam23@gmail.com) [www. nature.com/scientificreports/](www. nature.com/scientificreports/) õcal disorder that affects nearly 50 million people across the globe, interfering with their usual activities, thought processes, and socialization1. Despite improvements in therapeutic techniques and technology, most of the patients are still plagued by unpredictable seizures, no doubt indicating the requirement for developing accurate predicti

FAQs

What problems do existing seizure detection methods face?

They often show poor applicability across cases, produce excessive incorrect alerts, and rely on inefficient feature-selection procedures.

How does the proposed EpilepNet-LD framework improve epilepsy prediction?

It combines federated learning for privacy-preserving decentralized training with HSA-MI feature selection to identify effective temporal, spectral, and spatial EEG features.

What is the role of the EpilepNet-LD architecture in the system?

It uses LSTM to capture temporal dependencies and DenseNet-121 to extract hierarchical spatial features, improving seizure detection accuracy and sensitivity.

Sep 30, 2026

Uploaded by Melati