EEG-Based Seizure Detection Using Dual-Branch CNN-ViT Network Integrating Phase and Power Spectrograms

<b>Background/Objectives:</b> Epilepsy is a common neurological disorder with pathological mechanisms closely associated with the spatiotemporal dynamic characteristics of electroencephalogram (EEG) signals. Although significant progress has been made in epileptic seizure detection metho...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhuohan Wang, Yaoqi Hu, Qingyue Xin, Guanghao Jin, Yazhou Zhao, Weidong Zhou, Guoyang Liu
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Brain Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3425/15/5/509
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850127645935665152
author Zhuohan Wang
Yaoqi Hu
Qingyue Xin
Guanghao Jin
Yazhou Zhao
Weidong Zhou
Guoyang Liu
author_facet Zhuohan Wang
Yaoqi Hu
Qingyue Xin
Guanghao Jin
Yazhou Zhao
Weidong Zhou
Guoyang Liu
author_sort Zhuohan Wang
collection DOAJ
description <b>Background/Objectives:</b> Epilepsy is a common neurological disorder with pathological mechanisms closely associated with the spatiotemporal dynamic characteristics of electroencephalogram (EEG) signals. Although significant progress has been made in epileptic seizure detection methods using time–frequency analysis, current research still faces challenges in terms of an insufficient utilization of phase information. <b>Methods:</b> In this study, we propose an effective epileptic seizure detection framework based on continuous wavelet transform (CWT) and a hybrid network consisting of convolutional neural network (CNN) and vision transformer (ViT). First, the raw EEG signals are processed by the CWT. Then, the phase spectrogram and power spectrogram of the EEG are generated, and they are sent into the designed CNN and ViT branches of the network to extract more discriminative EEG features. Finally, the features output from the two branches are fused and fed into the classification network to obtain the detection results. <b>Results:</b> Experimental results on the CHB-MIT public dataset and our SH-SDU clinical dataset show that the proposed framework achieves sensitivities of 98.09% and 89.02%, specificities of 98.21% and 95.46%, and average accuracies of 98.45% and 94.66%, respectively. Furthermore, we compared the spectral characteristics of CWT with other time–frequency transforms within the hybrid architecture, demonstrating the advantages of the CWT-based CNN-ViT architecture. <b>Conclusions:</b> These results highlight the outstanding epileptic seizure detection performance of the proposed framework and its significant clinical feasibility.
format Article
id doaj-art-bb361de9365e4bf5852ab7b536064cbb
institution OA Journals
issn 2076-3425
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Brain Sciences
spelling doaj-art-bb361de9365e4bf5852ab7b536064cbb2025-08-20T02:33:38ZengMDPI AGBrain Sciences2076-34252025-05-0115550910.3390/brainsci15050509EEG-Based Seizure Detection Using Dual-Branch CNN-ViT Network Integrating Phase and Power SpectrogramsZhuohan Wang0Yaoqi Hu1Qingyue Xin2Guanghao Jin3Yazhou Zhao4Weidong Zhou5Guoyang Liu6School of Integrated Circuits, Shandong University, Jinan 250199, ChinaSchool of Integrated Circuits, Shandong University, Jinan 250199, ChinaSchool of Integrated Circuits, Shandong University, Jinan 250199, ChinaInstitute of Computer Science, Ludwig Maximilian University of Munich, 80539 Munich, GermanyDepartment of Biomedical Engineering, New York University, New York, NY 10012, USASchool of Integrated Circuits, Shandong University, Jinan 250199, ChinaSchool of Integrated Circuits, Shandong University, Jinan 250199, China<b>Background/Objectives:</b> Epilepsy is a common neurological disorder with pathological mechanisms closely associated with the spatiotemporal dynamic characteristics of electroencephalogram (EEG) signals. Although significant progress has been made in epileptic seizure detection methods using time–frequency analysis, current research still faces challenges in terms of an insufficient utilization of phase information. <b>Methods:</b> In this study, we propose an effective epileptic seizure detection framework based on continuous wavelet transform (CWT) and a hybrid network consisting of convolutional neural network (CNN) and vision transformer (ViT). First, the raw EEG signals are processed by the CWT. Then, the phase spectrogram and power spectrogram of the EEG are generated, and they are sent into the designed CNN and ViT branches of the network to extract more discriminative EEG features. Finally, the features output from the two branches are fused and fed into the classification network to obtain the detection results. <b>Results:</b> Experimental results on the CHB-MIT public dataset and our SH-SDU clinical dataset show that the proposed framework achieves sensitivities of 98.09% and 89.02%, specificities of 98.21% and 95.46%, and average accuracies of 98.45% and 94.66%, respectively. Furthermore, we compared the spectral characteristics of CWT with other time–frequency transforms within the hybrid architecture, demonstrating the advantages of the CWT-based CNN-ViT architecture. <b>Conclusions:</b> These results highlight the outstanding epileptic seizure detection performance of the proposed framework and its significant clinical feasibility.https://www.mdpi.com/2076-3425/15/5/509seizure detectioncontinuous wavelet transformconvolutional neural networkvision transformer
spellingShingle Zhuohan Wang
Yaoqi Hu
Qingyue Xin
Guanghao Jin
Yazhou Zhao
Weidong Zhou
Guoyang Liu
EEG-Based Seizure Detection Using Dual-Branch CNN-ViT Network Integrating Phase and Power Spectrograms
Brain Sciences
seizure detection
continuous wavelet transform
convolutional neural network
vision transformer
title EEG-Based Seizure Detection Using Dual-Branch CNN-ViT Network Integrating Phase and Power Spectrograms
title_full EEG-Based Seizure Detection Using Dual-Branch CNN-ViT Network Integrating Phase and Power Spectrograms
title_fullStr EEG-Based Seizure Detection Using Dual-Branch CNN-ViT Network Integrating Phase and Power Spectrograms
title_full_unstemmed EEG-Based Seizure Detection Using Dual-Branch CNN-ViT Network Integrating Phase and Power Spectrograms
title_short EEG-Based Seizure Detection Using Dual-Branch CNN-ViT Network Integrating Phase and Power Spectrograms
title_sort eeg based seizure detection using dual branch cnn vit network integrating phase and power spectrograms
topic seizure detection
continuous wavelet transform
convolutional neural network
vision transformer
url https://www.mdpi.com/2076-3425/15/5/509
work_keys_str_mv AT zhuohanwang eegbasedseizuredetectionusingdualbranchcnnvitnetworkintegratingphaseandpowerspectrograms
AT yaoqihu eegbasedseizuredetectionusingdualbranchcnnvitnetworkintegratingphaseandpowerspectrograms
AT qingyuexin eegbasedseizuredetectionusingdualbranchcnnvitnetworkintegratingphaseandpowerspectrograms
AT guanghaojin eegbasedseizuredetectionusingdualbranchcnnvitnetworkintegratingphaseandpowerspectrograms
AT yazhouzhao eegbasedseizuredetectionusingdualbranchcnnvitnetworkintegratingphaseandpowerspectrograms
AT weidongzhou eegbasedseizuredetectionusingdualbranchcnnvitnetworkintegratingphaseandpowerspectrograms
AT guoyangliu eegbasedseizuredetectionusingdualbranchcnnvitnetworkintegratingphaseandpowerspectrograms