An Attention-Enhanced 3D-CNN Framework for Spectrogram-Based EEG Analysis in Epilepsy Detection

Epilepsy is a widespread neurological disorder affecting approximately 50 million individuals globally, with a disproportionately high burden in low- and middle-income countries. It is characterized by recurrent seizures caused by sudden and uncontrolled electrical discharges in brain cells, often l...

Full description

Saved in:
Bibliographic Details
Main Authors: Ziaullah Khan, Aakanksha Dayal, Hee-Cheol Kim
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11017574/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849317999981363200
author Ziaullah Khan
Aakanksha Dayal
Hee-Cheol Kim
author_facet Ziaullah Khan
Aakanksha Dayal
Hee-Cheol Kim
author_sort Ziaullah Khan
collection DOAJ
description Epilepsy is a widespread neurological disorder affecting approximately 50 million individuals globally, with a disproportionately high burden in low- and middle-income countries. It is characterized by recurrent seizures caused by sudden and uncontrolled electrical discharges in brain cells, often leading to cognitive and motor impairments. Electroencephalography (EEG) remains the gold standard for epilepsy diagnosis, offering non-invasive monitoring of brain activity. However, the complexity and variability of epileptic patterns make traditional visual analysis subjective, time-consuming, and impractical for continuous monitoring. This underscores the need for automated deep learning solutions to improve diagnostic accuracy and operational efficiency. This study introduces a novel 3D Convolutional Neural Network with integrated Self-Attention (3D-ACNN), specifically designed for the volumetric analysis of multi-channel EEG data. The proposed algorithm preprocesses 21-channel EEG signals by segmenting them into 4-second Hann windows, applies the Short-Time Fourier Transform (STFT) to extract time-frequency features, and constructs stacked spectrograms that serve as three-dimensional inputs to the model. The proposed model (3D-ACNN), integrated with self-attention mechanisms, effectively captures spatiotemporal dependencies, enhancing seizure classification accuracy. The model achieved a classification accuracy of 99.89.
format Article
id doaj-art-77fb7b20a1f343cab5a053f092f835fa
institution Kabale University
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-77fb7b20a1f343cab5a053f092f835fa2025-08-20T03:51:03ZengIEEEIEEE Access2169-35362025-01-011312043712044510.1109/ACCESS.2025.357464611017574An Attention-Enhanced 3D-CNN Framework for Spectrogram-Based EEG Analysis in Epilepsy DetectionZiaullah Khan0https://orcid.org/0009-0008-3518-3712Aakanksha Dayal1https://orcid.org/0009-0005-4026-9615Hee-Cheol Kim2https://orcid.org/0000-0002-5399-7647Department of Digital Anti-Aging Healthcare, Inje University, Gimhae, South KoreaDepartment of Computer Engineering, Inje University, Gimhae, South KoreaDigital Anti-Aging Healthcare, College of AI Convergence, u-AHRC, Inje University, Gimhae, South KoreaEpilepsy is a widespread neurological disorder affecting approximately 50 million individuals globally, with a disproportionately high burden in low- and middle-income countries. It is characterized by recurrent seizures caused by sudden and uncontrolled electrical discharges in brain cells, often leading to cognitive and motor impairments. Electroencephalography (EEG) remains the gold standard for epilepsy diagnosis, offering non-invasive monitoring of brain activity. However, the complexity and variability of epileptic patterns make traditional visual analysis subjective, time-consuming, and impractical for continuous monitoring. This underscores the need for automated deep learning solutions to improve diagnostic accuracy and operational efficiency. This study introduces a novel 3D Convolutional Neural Network with integrated Self-Attention (3D-ACNN), specifically designed for the volumetric analysis of multi-channel EEG data. The proposed algorithm preprocesses 21-channel EEG signals by segmenting them into 4-second Hann windows, applies the Short-Time Fourier Transform (STFT) to extract time-frequency features, and constructs stacked spectrograms that serve as three-dimensional inputs to the model. The proposed model (3D-ACNN), integrated with self-attention mechanisms, effectively captures spatiotemporal dependencies, enhancing seizure classification accuracy. The model achieved a classification accuracy of 99.89.https://ieeexplore.ieee.org/document/11017574/EEG signal processing3D convolutional neural network (3D-CNN)biomedical signal analysisseizure detectionself-attention mechanismshort-time Fourier transform
spellingShingle Ziaullah Khan
Aakanksha Dayal
Hee-Cheol Kim
An Attention-Enhanced 3D-CNN Framework for Spectrogram-Based EEG Analysis in Epilepsy Detection
IEEE Access
EEG signal processing
3D convolutional neural network (3D-CNN)
biomedical signal analysis
seizure detection
self-attention mechanism
short-time Fourier transform
title An Attention-Enhanced 3D-CNN Framework for Spectrogram-Based EEG Analysis in Epilepsy Detection
title_full An Attention-Enhanced 3D-CNN Framework for Spectrogram-Based EEG Analysis in Epilepsy Detection
title_fullStr An Attention-Enhanced 3D-CNN Framework for Spectrogram-Based EEG Analysis in Epilepsy Detection
title_full_unstemmed An Attention-Enhanced 3D-CNN Framework for Spectrogram-Based EEG Analysis in Epilepsy Detection
title_short An Attention-Enhanced 3D-CNN Framework for Spectrogram-Based EEG Analysis in Epilepsy Detection
title_sort attention enhanced 3d cnn framework for spectrogram based eeg analysis in epilepsy detection
topic EEG signal processing
3D convolutional neural network (3D-CNN)
biomedical signal analysis
seizure detection
self-attention mechanism
short-time Fourier transform
url https://ieeexplore.ieee.org/document/11017574/
work_keys_str_mv AT ziaullahkhan anattentionenhanced3dcnnframeworkforspectrogrambasedeeganalysisinepilepsydetection
AT aakankshadayal anattentionenhanced3dcnnframeworkforspectrogrambasedeeganalysisinepilepsydetection
AT heecheolkim anattentionenhanced3dcnnframeworkforspectrogrambasedeeganalysisinepilepsydetection
AT ziaullahkhan attentionenhanced3dcnnframeworkforspectrogrambasedeeganalysisinepilepsydetection
AT aakankshadayal attentionenhanced3dcnnframeworkforspectrogrambasedeeganalysisinepilepsydetection
AT heecheolkim attentionenhanced3dcnnframeworkforspectrogrambasedeeganalysisinepilepsydetection