A novel EEG artifact removal algorithm based on an advanced attention mechanism

Abstract EEG is widely applied in emotion recognition, brain disease detection, and other fields due to its high temporal resolution and non-invasiveness. However, artifact removal remains a crucial issue in EEG signal processing. Recently, with the rapid development of deep learning, there has been...

Full description

Saved in:
Bibliographic Details
Main Authors: Rui Jiang, Shen Tong, Jiawei Wu, Haowei Hu, Ran Zhang, Heng Wang, Yan Zhao, Weixin Zhu, Shuyan Li, Xiao Zhang
Format: Article
Language:English
Published: Nature Portfolio 2025-06-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-98653-1
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850223903093293056
author Rui Jiang
Shen Tong
Jiawei Wu
Haowei Hu
Ran Zhang
Heng Wang
Yan Zhao
Weixin Zhu
Shuyan Li
Xiao Zhang
author_facet Rui Jiang
Shen Tong
Jiawei Wu
Haowei Hu
Ran Zhang
Heng Wang
Yan Zhao
Weixin Zhu
Shuyan Li
Xiao Zhang
author_sort Rui Jiang
collection DOAJ
description Abstract EEG is widely applied in emotion recognition, brain disease detection, and other fields due to its high temporal resolution and non-invasiveness. However, artifact removal remains a crucial issue in EEG signal processing. Recently, with the rapid development of deep learning, there has been a significant transformation in the methods of EEG artifact removal. Nonetheless, existing research still exhibits some limitations: (1) insufficient capability to remove unknown artifacts; (2) inability to adapt to tasks where artifact removal needs to be applied to the overall input of multi-channel EEG data. Therefore, this study proposes CLEnet by integrating dual-scale CNN (Convolutional Neural Networks) and LSTM (Long Short-Term Memory), and incorporating an improved EMA-1D (One-Dimensional Efficient Multi-Scale Attention Mechanism). CLEnet can extract the morphological features and temporal features of EEG, thereby separating EEG from artifacts. We conducted experiments on three datasets, and the results showed that CLEnet performed best. Specifically, in the task of removing artifacts from multi-channel EEG data containing unknown artifacts, CLEnet shows improvements of 2.45% and 2.65% in SNR(signal-to-noise ratio) and CC(average correlation coefficient). Moreover, RRMSE t (relative root mean square error in the temporal domain) and RRMSE f (relative root mean square error in the frequency domain) decrease by 6.94% and 3.30%.
format Article
id doaj-art-d576a6e2974c42a48e458128349877fa
institution OA Journals
issn 2045-2322
language English
publishDate 2025-06-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-d576a6e2974c42a48e458128349877fa2025-08-20T02:05:48ZengNature PortfolioScientific Reports2045-23222025-06-0115111910.1038/s41598-025-98653-1A novel EEG artifact removal algorithm based on an advanced attention mechanismRui Jiang0Shen Tong1Jiawei Wu2Haowei Hu3Ran Zhang4Heng Wang5Yan Zhao6Weixin Zhu7Shuyan Li8Xiao Zhang9School of Medical Information and Engineering, Xuzhou Medical UniversitySchool of Medical Information and Engineering, Xuzhou Medical UniversitySchool of Medical Information and Engineering, Xuzhou Medical UniversitySchool of Medical Information and Engineering, Xuzhou Medical UniversitySchool of Medical Information and Engineering, Xuzhou Medical UniversitySchool of Medical Information and Engineering, Xuzhou Medical UniversityXuzhou Jiazhi information Technology Co., Ltd.Jinhua Central HospitalSchool of Medical Information and Engineering, Xuzhou Medical UniversitySchool of Medicine and Health, Harbin Institute of TechnologyAbstract EEG is widely applied in emotion recognition, brain disease detection, and other fields due to its high temporal resolution and non-invasiveness. However, artifact removal remains a crucial issue in EEG signal processing. Recently, with the rapid development of deep learning, there has been a significant transformation in the methods of EEG artifact removal. Nonetheless, existing research still exhibits some limitations: (1) insufficient capability to remove unknown artifacts; (2) inability to adapt to tasks where artifact removal needs to be applied to the overall input of multi-channel EEG data. Therefore, this study proposes CLEnet by integrating dual-scale CNN (Convolutional Neural Networks) and LSTM (Long Short-Term Memory), and incorporating an improved EMA-1D (One-Dimensional Efficient Multi-Scale Attention Mechanism). CLEnet can extract the morphological features and temporal features of EEG, thereby separating EEG from artifacts. We conducted experiments on three datasets, and the results showed that CLEnet performed best. Specifically, in the task of removing artifacts from multi-channel EEG data containing unknown artifacts, CLEnet shows improvements of 2.45% and 2.65% in SNR(signal-to-noise ratio) and CC(average correlation coefficient). Moreover, RRMSE t (relative root mean square error in the temporal domain) and RRMSE f (relative root mean square error in the frequency domain) decrease by 6.94% and 3.30%.https://doi.org/10.1038/s41598-025-98653-1Electroencephalogram (EEG)Artifact removalDeep learningAttention mechanism
spellingShingle Rui Jiang
Shen Tong
Jiawei Wu
Haowei Hu
Ran Zhang
Heng Wang
Yan Zhao
Weixin Zhu
Shuyan Li
Xiao Zhang
A novel EEG artifact removal algorithm based on an advanced attention mechanism
Scientific Reports
Electroencephalogram (EEG)
Artifact removal
Deep learning
Attention mechanism
title A novel EEG artifact removal algorithm based on an advanced attention mechanism
title_full A novel EEG artifact removal algorithm based on an advanced attention mechanism
title_fullStr A novel EEG artifact removal algorithm based on an advanced attention mechanism
title_full_unstemmed A novel EEG artifact removal algorithm based on an advanced attention mechanism
title_short A novel EEG artifact removal algorithm based on an advanced attention mechanism
title_sort novel eeg artifact removal algorithm based on an advanced attention mechanism
topic Electroencephalogram (EEG)
Artifact removal
Deep learning
Attention mechanism
url https://doi.org/10.1038/s41598-025-98653-1
work_keys_str_mv AT ruijiang anoveleegartifactremovalalgorithmbasedonanadvancedattentionmechanism
AT shentong anoveleegartifactremovalalgorithmbasedonanadvancedattentionmechanism
AT jiaweiwu anoveleegartifactremovalalgorithmbasedonanadvancedattentionmechanism
AT haoweihu anoveleegartifactremovalalgorithmbasedonanadvancedattentionmechanism
AT ranzhang anoveleegartifactremovalalgorithmbasedonanadvancedattentionmechanism
AT hengwang anoveleegartifactremovalalgorithmbasedonanadvancedattentionmechanism
AT yanzhao anoveleegartifactremovalalgorithmbasedonanadvancedattentionmechanism
AT weixinzhu anoveleegartifactremovalalgorithmbasedonanadvancedattentionmechanism
AT shuyanli anoveleegartifactremovalalgorithmbasedonanadvancedattentionmechanism
AT xiaozhang anoveleegartifactremovalalgorithmbasedonanadvancedattentionmechanism
AT ruijiang noveleegartifactremovalalgorithmbasedonanadvancedattentionmechanism
AT shentong noveleegartifactremovalalgorithmbasedonanadvancedattentionmechanism
AT jiaweiwu noveleegartifactremovalalgorithmbasedonanadvancedattentionmechanism
AT haoweihu noveleegartifactremovalalgorithmbasedonanadvancedattentionmechanism
AT ranzhang noveleegartifactremovalalgorithmbasedonanadvancedattentionmechanism
AT hengwang noveleegartifactremovalalgorithmbasedonanadvancedattentionmechanism
AT yanzhao noveleegartifactremovalalgorithmbasedonanadvancedattentionmechanism
AT weixinzhu noveleegartifactremovalalgorithmbasedonanadvancedattentionmechanism
AT shuyanli noveleegartifactremovalalgorithmbasedonanadvancedattentionmechanism
AT xiaozhang noveleegartifactremovalalgorithmbasedonanadvancedattentionmechanism