View in EDS

Cycle-Consistent Simplicial Adversarial Adaptation Network with Spider Wasp Optimizer-Based EEG Signals Classification Model for Epileptic Seizure.

Saved in:
Bibliographic Details
Title: Cycle-Consistent Simplicial Adversarial Adaptation Network with Spider Wasp Optimizer-Based EEG Signals Classification Model for Epileptic Seizure.
Authors: Singh, Law Kumar1 (AUTHOR) Lawkumarcs1@gmail.com, Mathivanan, M.2 (AUTHOR) mathivananacs@gmail.com, Ananthi, P.3 (AUTHOR) P.1.4Ananthi@outlook.com, Sitaraman, Surendar Rama4 (AUTHOR) surendar.rama.sitaraman@ieee.org
Source: Circuits, Systems & Signal Processing. Dec2025, Vol. 44 Issue 12, p9512-9548. 37p.
Subjects: Epilepsy, Classification, Feature extraction, Optimization algorithms, Neurophysiology, Fokker-Planck equation, Curve fitting
Abstract: EEG analysis of brainwave patterns is used in EEG signal categorization for epileptic seizures (CES) to detect abnormal activity suggestive of seizures. Classifying EEG signals for epileptic seizures (ES) is essential for early detection and management; however, due to the non-stationary nature, noisy characteristics, and inter-patient variability of EEGs, choosing a reliable and long-lasting approach for this task is difficult. For the purpose of successfully identifying the ES EEG signals, this study proposes a Cycle-consistent Simplicial Adversarial Adaptation Network with Spider Wasp Optimizer (C-CSAA2Nets + SWO) framework. The TUSZ and CHB-MIT databases provided the raw EEG signals used in this study. These raw EEG signals undergo preprocessing through the Cumulative Curve Fitting Approximation (CCFA) algorithm to enhance signal quality, denoising the signals, artifact removal, and noise reduction. Following preprocessing, feature extraction is performed using the Fokker–Planck Equations using Laguerre Wavelet Transform (F-PELWT). Following optimization using the Spider Wasp Optimizer (SWO), these extracted features are classified using the Cycle-consistent Simplicial Adversarial Adaptation Network (C-CSAA2Nets). We implement the proposed C-CSAA2Nets + SWO paradigm in Python. When tested on the TUSZ and CHB-MIT datasets, the suggested approach outperformed existing techniques with an impressive accuracy of 99.9% and sensitivity of 98.9%. The CES from EEG signals is greatly enhanced by the results of the suggested approach. By leveraging domain adaptation and optimizing classification, the model achieves higher classification accuracy and generalization, offering a more reliable tool for ES detection in clinical applications. [ABSTRACT FROM AUTHOR]
Copyright of Circuits, Systems & Signal Processing is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
Description
Abstract:EEG analysis of brainwave patterns is used in EEG signal categorization for epileptic seizures (CES) to detect abnormal activity suggestive of seizures. Classifying EEG signals for epileptic seizures (ES) is essential for early detection and management; however, due to the non-stationary nature, noisy characteristics, and inter-patient variability of EEGs, choosing a reliable and long-lasting approach for this task is difficult. For the purpose of successfully identifying the ES EEG signals, this study proposes a Cycle-consistent Simplicial Adversarial Adaptation Network with Spider Wasp Optimizer (C-CSAA2Nets + SWO) framework. The TUSZ and CHB-MIT databases provided the raw EEG signals used in this study. These raw EEG signals undergo preprocessing through the Cumulative Curve Fitting Approximation (CCFA) algorithm to enhance signal quality, denoising the signals, artifact removal, and noise reduction. Following preprocessing, feature extraction is performed using the Fokker–Planck Equations using Laguerre Wavelet Transform (F-PELWT). Following optimization using the Spider Wasp Optimizer (SWO), these extracted features are classified using the Cycle-consistent Simplicial Adversarial Adaptation Network (C-CSAA2Nets). We implement the proposed C-CSAA2Nets + SWO paradigm in Python. When tested on the TUSZ and CHB-MIT datasets, the suggested approach outperformed existing techniques with an impressive accuracy of 99.9% and sensitivity of 98.9%. The CES from EEG signals is greatly enhanced by the results of the suggested approach. By leveraging domain adaptation and optimizing classification, the model achieves higher classification accuracy and generalization, offering a more reliable tool for ES detection in clinical applications. [ABSTRACT FROM AUTHOR]
ISSN:0278081X
DOI:10.1007/s00034-025-03250-2