Epileptic Seizure Detection Contribution in Healthcare Sustainability

Title

Epileptic Seizure Detection Contribution in Healthcare Sustainability

Creator

Kumar, Basant; Najmusseher; Nizar Banu, P.K.; Dwivedi, Rashmi

Description

This study describes a sustainable EEG data methodology. Classification using Discrete Wavelet Transform (DWT) for feature extraction, with the objective of reducing the computational efforts while keeping accurate neural signal analysis. DWT decomposes the EEG signal into timefrequency specific components which allows extraction of ten key wavelet features, including wavelet energy, entropy, maximal coefficients, zero-crossing counts, and dominant frequency. These features capture essential timefrequency features of EEG signals, providing a comprehensive yet computationally efficient representation. By streamlining feature extraction, this approach reduces data dimensionality and minimizes computational processing time, aligning with sustainable technology objectives. The resulting feature vectors serve as robust inputs for classification models, effectively supporting EEG data interpretation with reduced energy and less resource utilization. This study demonstrates that targeted feature extraction can achieve high classification performance in EEG analysis while adhering to principles of sustainability and resource efficiency. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

Source

Studies in Systems, Decision and Control;Volume;601;pp.225-235

Date

01-01-2025

Publisher

Springer Science and Business Media Deutschland GmbH

Subject

EEG; Epilepsy; Seizures; Signal transformation; Wavelet

Coverage

Kumar B., Modern College of Business and Science, Muscat, Oman; Najmusseher, Department of Computer Science, CHRIST University Central Campus, Karnataka, Bangalore, India; Nizar Banu P.K., Department of Computer Science, CHRIST University Central Campus, Karnataka, Bangalore, India; Dwivedi R., Muscat University, Muscat, Oman

Rights

Restricted Access; Hardcopy may be available in the library

Relation

ISSN: 21984182;

Format

online

Language

English

Type

Book chapter

Identifier

https://doi.org/10.1007/978-3-031-92240-4_21

https://www.scopus.com/pages/publications/105014173618?origin=resultslist