DMAE-EEG: A Pretraining Framework for EEG Spatiotemporal Representation Learning

Electroencephalography (EEG) plays a crucial role in neuroscience research and clinical practice, but it remains limited by nonuniform data, noise, and difficulty in labeling. To address these challenges, we develop a pretraining framework named DMAE-EEG, a denoising masked autoencoder for mining ge...

Full description

Saved in:

Bibliographic Details
Published in:	IEEE transaction on neural networks and learning systems Vol. 36; no. 10; pp. 17664 - 17678
Main Authors:	Zhang, Yifan, Yu, Yang, Li, Hao, Wu, Anqi, Chen, Xin, Liu, Jinfang, Zeng, Ling-Li, Hu, Dewen
Format:	Journal Article
Language:	English
Published:	United States IEEE 01.10.2025
Subjects:	Algorithms Autoencoders Brain - physiology Brain modeling Databases, Factual Decoding Electrodes Electroencephalography Electroencephalography (EEG) Electroencephalography - methods Feature extraction Humans Machine Learning masked autoencoder motion intention recognition Motors Neural Networks, Computer Noise Representation learning Signal Processing, Computer-Assisted signal quality enhancement Signal-To-Noise Ratio Spatio-Temporal Analysis Spatiotemporal phenomena
ISSN:	2162-237X, 2162-2388, 2162-2388
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Abstract	Electroencephalography (EEG) plays a crucial role in neuroscience research and clinical practice, but it remains limited by nonuniform data, noise, and difficulty in labeling. To address these challenges, we develop a pretraining framework named DMAE-EEG, a denoising masked autoencoder for mining generalizable spatiotemporal representation from massive unlabeled EEG. First, we propose a novel brain region topological heterogeneity (BRTH) division method to partition the nonuniform data into fixed patches based on neuroscientific priors. Second, we design a denoised pseudo-label generator (DPLG), which utilizes a denoising reconstruction pretext task to enable the learning of generalizable representations from massive unlabeled EEG, suppressing the influence of noise and artifacts. Furthermore, we utilize an asymmetric autoencoder with self-attention as the backbone in the proposed DMAE-EEG, which captures long-range spatiotemporal dependencies and interactions from unlabeled EEG data across 14 public datasets. The proposed DMAE-EEG is validated on both generative (signal quality enhancement) and discriminative tasks (motion intention recognition). In the quality enhancement, DMAE-EEG outperforms existing statistical methods with normalized mean squared error (nMSE) reduction of 27.78%-50.00% under corruption levels of 25%, 50%, and 75%, respectively. In motion intention recognition, DMAE-EEG achieves a relative improvement of 2.71%-6.14% in intrasession classification balanced accuracy across 2-6 class motor execution and imagery tasks, outperforming state-of-the-art methods. Overall, the results suggest that the pretraining framework DMAE-EEG can capture generalizable spatiotemporal representations from massive unlabeled EEG and enhance the knowledge transferability across sessions, subjects, and tasks in various downstream scenarios, advancing EEG-aided diagnosis and brain-computer communication and control, and other clinical practice.
AbstractList	Electroencephalography (EEG) plays a crucial role in neuroscience research and clinical practice, but it remains limited by nonuniform data, noise, and difficulty in labeling. To address these challenges, we develop a pretraining framework named DMAE-EEG, a denoising masked autoencoder for mining generalizable spatiotemporal representation from massive unlabeled EEG. First, we propose a novel brain region topological heterogeneity (BRTH) division method to partition the nonuniform data into fixed patches based on neuroscientific priors. Second, we design a denoised pseudo-label generator (DPLG), which utilizes a denoising reconstruction pretext task to enable the learning of generalizable representations from massive unlabeled EEG, suppressing the influence of noise and artifacts. Furthermore, we utilize an asymmetric autoencoder with self-attention as the backbone in the proposed DMAE-EEG, which captures long-range spatiotemporal dependencies and interactions from unlabeled EEG data across 14 public datasets. The proposed DMAE-EEG is validated on both generative (signal quality enhancement) and discriminative tasks (motion intention recognition). In the quality enhancement, DMAE-EEG outperforms existing statistical methods with normalized mean squared error (nMSE) reduction of 27.78%-50.00% under corruption levels of 25%, 50%, and 75%, respectively. In motion intention recognition, DMAE-EEG achieves a relative improvement of 2.71%-6.14% in intrasession classification balanced accuracy across 2-6 class motor execution and imagery tasks, outperforming state-of-the-art methods. Overall, the results suggest that the pretraining framework DMAE-EEG can capture generalizable spatiotemporal representations from massive unlabeled EEG and enhance the knowledge transferability across sessions, subjects, and tasks in various downstream scenarios, advancing EEG-aided diagnosis and brain-computer communication and control, and other clinical practice. Electroencephalography (EEG) plays a crucial role in neuroscience research and clinical practice, but it remains limited by nonuniform data, noise, and difficulty in labeling. To address these challenges, we develop a pretraining framework named DMAE-EEG, a denoising masked autoencoder for mining generalizable spatiotemporal representation from massive unlabeled EEG. First, we propose a novel brain region topological heterogeneity (BRTH) division method to partition the nonuniform data into fixed patches based on neuroscientific priors. Second, we design a denoised pseudo-label generator (DPLG), which utilizes a denoising reconstruction pretext task to enable the learning of generalizable representations from massive unlabeled EEG, suppressing the influence of noise and artifacts. Furthermore, we utilize an asymmetric autoencoder with self-attention as the backbone in the proposed DMAE-EEG, which captures long-range spatiotemporal dependencies and interactions from unlabeled EEG data across 14 public datasets. The proposed DMAE-EEG is validated on both generative (signal quality enhancement) and discriminative tasks (motion intention recognition). In the quality enhancement, DMAE-EEG outperforms existing statistical methods with normalized mean squared error (nMSE) reduction of 27.78%-50.00% under corruption levels of 25%, 50%, and 75%, respectively. In motion intention recognition, DMAE-EEG achieves a relative improvement of 2.71%-6.14% in intrasession classification balanced accuracy across 2-6 class motor execution and imagery tasks, outperforming state-of-the-art methods. Overall, the results suggest that the pretraining framework DMAE-EEG can capture generalizable spatiotemporal representations from massive unlabeled EEG and enhance the knowledge transferability across sessions, subjects, and tasks in various downstream scenarios, advancing EEG-aided diagnosis and brain-computer communication and control, and other clinical practice.Electroencephalography (EEG) plays a crucial role in neuroscience research and clinical practice, but it remains limited by nonuniform data, noise, and difficulty in labeling. To address these challenges, we develop a pretraining framework named DMAE-EEG, a denoising masked autoencoder for mining generalizable spatiotemporal representation from massive unlabeled EEG. First, we propose a novel brain region topological heterogeneity (BRTH) division method to partition the nonuniform data into fixed patches based on neuroscientific priors. Second, we design a denoised pseudo-label generator (DPLG), which utilizes a denoising reconstruction pretext task to enable the learning of generalizable representations from massive unlabeled EEG, suppressing the influence of noise and artifacts. Furthermore, we utilize an asymmetric autoencoder with self-attention as the backbone in the proposed DMAE-EEG, which captures long-range spatiotemporal dependencies and interactions from unlabeled EEG data across 14 public datasets. The proposed DMAE-EEG is validated on both generative (signal quality enhancement) and discriminative tasks (motion intention recognition). In the quality enhancement, DMAE-EEG outperforms existing statistical methods with normalized mean squared error (nMSE) reduction of 27.78%-50.00% under corruption levels of 25%, 50%, and 75%, respectively. In motion intention recognition, DMAE-EEG achieves a relative improvement of 2.71%-6.14% in intrasession classification balanced accuracy across 2-6 class motor execution and imagery tasks, outperforming state-of-the-art methods. Overall, the results suggest that the pretraining framework DMAE-EEG can capture generalizable spatiotemporal representations from massive unlabeled EEG and enhance the knowledge transferability across sessions, subjects, and tasks in various downstream scenarios, advancing EEG-aided diagnosis and brain-computer communication and control, and other clinical practice.
Author	Wu, Anqi Yu, Yang Chen, Xin Zhang, Yifan Li, Hao Zeng, Ling-Li Liu, Jinfang Hu, Dewen
Author_xml	– sequence: 1 givenname: Yifan orcidid: 0000-0002-4671-723X surname: Zhang fullname: Zhang, Yifan organization: College of Intelligence Science and Technology, National University of Defense Technology, Changsha, China – sequence: 2 givenname: Yang orcidid: 0000-0002-8967-0427 surname: Yu fullname: Yu, Yang organization: College of Intelligence Science and Technology, National University of Defense Technology, Changsha, China – sequence: 3 givenname: Hao orcidid: 0000-0002-9542-787X surname: Li fullname: Li, Hao organization: College of Intelligence Science and Technology, National University of Defense Technology, Changsha, China – sequence: 4 givenname: Anqi surname: Wu fullname: Wu, Anqi organization: College of Intelligence Science and Technology, National University of Defense Technology, Changsha, China – sequence: 5 givenname: Xin orcidid: 0000-0002-0672-0207 surname: Chen fullname: Chen, Xin organization: Department of Neurosurgery, Xiangya Hospital, National Clinical Medical Research Center for Geriatric Diseases, Central South University, Changsha, China – sequence: 6 givenname: Jinfang orcidid: 0000-0001-6986-1173 surname: Liu fullname: Liu, Jinfang organization: Department of Neurosurgery, Xiangya Hospital, National Clinical Medical Research Center for Geriatric Diseases, Central South University, Changsha, China – sequence: 7 givenname: Ling-Li orcidid: 0000-0002-0515-256X surname: Zeng fullname: Zeng, Ling-Li email: zengphd@nudt.edu.cn organization: College of Intelligence Science and Technology, National University of Defense Technology, Changsha, China – sequence: 8 givenname: Dewen orcidid: 0000-0001-7357-0053 surname: Hu fullname: Hu, Dewen email: dwhu@nudt.edu.cn organization: College of Intelligence Science and Technology, National University of Defense Technology, Changsha, China
BackLink	https://www.ncbi.nlm.nih.gov/pubmed/40601454$$D View this record in MEDLINE/PubMed
BookMark	eNpFkE1PwkAQhjcGI4j8AWNMj16K-71dbwQLmiASwcRbs7Szpko_3C0x_nuLoM5l3kyedw7PKeqUVQkInRM8JATr69V8PlsOKaZiyEREtCZHqEeJpCFlUdT5y-qliwbev-F2JBaS6xPU5W0kXPAeWtw-jOIwjqc3wShYOGicycu8fA0mzhTwWbn3wFYuaIFgWZsmrxoo6sqZTfAEtQMPZbO7lsEMjNsVz9CxNRsPg8Puo-dJvBrfhbPH6f14NAtTqlQTMgMpaGZJZFOiuFbCCLBWRVryVJOMg-KZ4oSnlnGss8hKkKnBAmewXgvN-uhq_7d21ccWfJMUuU9hszElVFufMEqlIkwI3qKXB3S7LiBLapcXxn0lvxZagO6B1FXeO7B_CMHJznbyYzvZ2U4OttvSxb6UA8B_gWBJtZLsGzi1ebY
CODEN	ITNNAL
Cites_doi	10.3390/app11041380 10.1007/s11023-020-09548-1 10.1145/3577925 10.1109/CVPR46437.2021.01212 10.1109/TPAMI.2014.2330598 10.1007/s11571-020-09634-1 10.1016/j.jneumeth.2020.108833 10.1109/EMBC44109.2020.9175874 10.1038/nrneurol.2016.113 10.1097/00004691-198804000-00005 10.1088/1741-2552/ac23e2 10.1007/978-3-658-40442-0_9 10.1016/s1388-2457(00)00527-7 10.1161/01.CTR.101.23.e215 10.1038/nrneurol.2012.150 10.1016/S0167-8760(00)00075-1 10.1016/j.neunet.2020.05.032 10.48550/ARXIV.1706.03762 10.1088/1741-2552/aace8c 10.1109/CVPR52688.2022.01553 10.1016/j.inffus.2023.03.022 10.1016/j.patcog.2022.108757 10.1093/gigascience/giaa098 10.1109/TNNLS.2020.3015505 10.1109/TBME.2004.827072 10.3389/fnins.2012.00039 10.1109/TNSRE.2022.3230250 10.1016/0013-4694(91)90202-F 10.1038/s42256-019-0091-7 10.1038/s41551-022-00914-1 10.1162/089976602317318938 10.1007/s11042-015-2717-z 10.1109/TAFFC.2022.3164516 10.1126/science.1127647 10.3389/fneur.2019.00325 10.1109/TPAMI.2022.3152247 10.1038/s42256-023-00714-5 10.1590/1980-57642016dn11-010002 10.1016/j.neuroimage.2021.118721 10.24003/emitter.v5i1.165 10.1109/JBHI.2023.3335854 10.1109/CCMB.2013.6609178 10.1111/j.1528-1167.2006.00655.x 10.1038/s41467-024-45922-8 10.1109/EMBC46164.2021.9629837 10.1109/CVPR42600.2020.00674
ContentType	Journal Article
DBID	97E RIA RIE AAYXX CITATION CGR CUY CVF ECM EIF NPM 7X8
DOI	10.1109/TNNLS.2025.3581991
DatabaseName	IEEE Xplore (IEEE) IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE Electronic Library (IEL) CrossRef Medline MEDLINE MEDLINE (Ovid) MEDLINE MEDLINE PubMed MEDLINE - Academic
DatabaseTitle	CrossRef MEDLINE Medline Complete MEDLINE with Full Text PubMed MEDLINE (Ovid) MEDLINE - Academic
DatabaseTitleList	MEDLINE MEDLINE - Academic
Database_xml	– sequence: 1 dbid: NPM name: PubMed url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 2 dbid: RIE name: IEEE Electronic Library (IEL) url: https://ieeexplore.ieee.org/ sourceTypes: Publisher – sequence: 3 dbid: 7X8 name: MEDLINE - Academic url: https://search.proquest.com/medline sourceTypes: Aggregation Database
DeliveryMethod	fulltext_linktorsrc
Discipline	Computer Science
EISSN	2162-2388
EndPage	17678
ExternalDocumentID	40601454 10_1109_TNNLS_2025_3581991 11062976
Genre	orig-research Journal Article
GrantInformation_xml	– fundername: Science and Technology Innovation Program of Hunan Province grantid: 2023RC1004; 2024QK2006 funderid: 10.13039/501100019081 – fundername: STI 2030—Major Projects grantid: 2022ZD0208903 – fundername: National Natural Science Foundation of China grantid: U24A20339; 62036013 funderid: 10.13039/501100001809
GroupedDBID	0R~ 4.4 5VS 6IK 97E AAJGR AASAJ AAWTH ABAZT ABQJQ ABVLG ACIWK ACPRK AENEX AFRAH AGQYO AGSQL AHBIQ AKJIK AKQYR ALMA_UNASSIGNED_HOLDINGS ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ EBS EJD IFIPE IPLJI JAVBF M43 MS~ O9- OCL PQQKQ RIA RIE RNS AAYXX CITATION CGR CUY CVF ECM EIF NPM 7X8
ID	FETCH-LOGICAL-c277t-3aece93f18fc174975a5eff78964c91d4e74d7414cf3409d8f6e6ca050debb593
IEDL.DBID	RIE
ISICitedReferencesCount	1
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001522955400001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	2162-237X 2162-2388
IngestDate	Thu Oct 02 22:28:33 EDT 2025 Thu Oct 09 01:30:56 EDT 2025 Sat Nov 29 07:11:57 EST 2025 Wed Oct 15 14:20:22 EDT 2025
IsPeerReviewed	false
IsScholarly	true
Issue	10
Language	English
License	https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html https://doi.org/10.15223/policy-029 https://doi.org/10.15223/policy-037
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c277t-3aece93f18fc174975a5eff78964c91d4e74d7414cf3409d8f6e6ca050debb593
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23
ORCID	0000-0002-8967-0427 0000-0002-9542-787X 0000-0002-4671-723X 0000-0001-7357-0053 0000-0002-0515-256X 0000-0002-0672-0207 0000-0001-6986-1173
PMID	40601454
PQID	3226713554
PQPubID	23479
PageCount	15
ParticipantIDs	ieee_primary_11062976 proquest_miscellaneous_3226713554 crossref_primary_10_1109_TNNLS_2025_3581991 pubmed_primary_40601454
PublicationCentury	2000
PublicationDate	2025-10-01
PublicationDateYYYYMMDD	2025-10-01
PublicationDate_xml	– month: 10 year: 2025 text: 2025-10-01 day: 01
PublicationDecade	2020
PublicationPlace	United States
PublicationPlace_xml	– name: United States
PublicationTitle	IEEE transaction on neural networks and learning systems
PublicationTitleAbbrev	TNNLS
PublicationTitleAlternate	IEEE Trans Neural Netw Learn Syst
PublicationYear	2025
Publisher	IEEE
Publisher_xml	– name: IEEE
References	ref13 ref12 Teplan (ref1) 2002; 2 ref15 Jasper (ref22) 1958; 10 ref14 Alsentzer (ref26) ref11 ref10 ref54 ref17 ref19 ref18 Jiang (ref53) ref50 ref46 ref45 ref48 ref47 ref41 ref44 ref43 ref49 ref8 ref7 ref9 ref4 Yang (ref51); 36 ref3 ref6 ref5 ref40 ref35 ref34 ref37 ref36 ref31 ref30 ref33 ref32 ref2 ref39 ref38 Dosovitskiy (ref42) ref24 ref23 ref25 ref20 Devlin (ref16) 2018; 1 ref21 Wang (ref52) ref28 ref27 ref29
References_xml	– ident: ref30 doi: 10.3390/app11041380 – ident: ref17 doi: 10.1007/s11023-020-09548-1 – ident: ref18 doi: 10.1145/3577925 – ident: ref40 doi: 10.1109/CVPR46437.2021.01212 – ident: ref47 doi: 10.1109/TPAMI.2014.2330598 – start-page: 238 volume-title: Proc. Mach. Learn. Health Workshop (ML4H) ident: ref26 article-title: Contrastive representation learning for electroencephalogram classification – ident: ref8 doi: 10.1007/s11571-020-09634-1 – ident: ref48 doi: 10.1016/j.jneumeth.2020.108833 – volume: 36 start-page: 32039 volume-title: Proc. Adv. Neural Inf. Process. Syst. ident: ref51 article-title: BIOT: Biosignal transformer for cross-data learning in the wild – ident: ref49 doi: 10.1109/EMBC44109.2020.9175874 – ident: ref2 doi: 10.1038/nrneurol.2016.113 – ident: ref23 doi: 10.1097/00004691-198804000-00005 – volume: 10 start-page: 371 year: 1958 ident: ref22 article-title: Ten-twenty electrode system of the international federation publication-title: Electroencephalogr. Clin. Neurophysiol. – ident: ref54 doi: 10.1088/1741-2552/ac23e2 – ident: ref29 doi: 10.1007/978-3-658-40442-0_9 – ident: ref24 doi: 10.1016/s1388-2457(00)00527-7 – ident: ref45 doi: 10.1161/01.CTR.101.23.e215 – ident: ref4 doi: 10.1038/nrneurol.2012.150 – ident: ref34 doi: 10.1016/S0167-8760(00)00075-1 – ident: ref39 doi: 10.1016/j.neunet.2020.05.032 – ident: ref44 doi: 10.48550/ARXIV.1706.03762 – ident: ref7 doi: 10.1088/1741-2552/aace8c – ident: ref43 doi: 10.1109/CVPR52688.2022.01553 – volume: 2 start-page: 1 issue: 2 year: 2002 ident: ref1 article-title: Fundamentals of EEG measurement publication-title: Meas. Sci. Technol. – ident: ref3 doi: 10.1016/j.inffus.2023.03.022 – ident: ref12 doi: 10.1016/j.patcog.2022.108757 – ident: ref46 doi: 10.1093/gigascience/giaa098 – ident: ref38 doi: 10.1109/TNNLS.2020.3015505 – ident: ref31 doi: 10.1109/TBME.2004.827072 – ident: ref37 doi: 10.3389/fnins.2012.00039 – start-page: 1 volume-title: Proc. 12th Int. Conf. Learn. Represent. ident: ref53 article-title: Large brain model for learning generic representations with tremendous EEG data in BCI – ident: ref11 doi: 10.1109/TNSRE.2022.3230250 – ident: ref13 doi: 10.1016/0013-4694(91)90202-F – ident: ref6 doi: 10.1038/s42256-019-0091-7 – start-page: 39249 volume-title: Proc. Adv. Neural Inf. Process. Syst. ident: ref52 article-title: EEGPT: Pretrained transformer for universal and reliable representation of EEG signals – ident: ref19 doi: 10.1038/s41551-022-00914-1 – ident: ref27 doi: 10.1162/089976602317318938 – ident: ref32 doi: 10.1007/s11042-015-2717-z – ident: ref21 doi: 10.1109/TAFFC.2022.3164516 – ident: ref28 doi: 10.1126/science.1127647 – ident: ref25 doi: 10.3389/fneur.2019.00325 – ident: ref41 doi: 10.1109/TPAMI.2022.3152247 – ident: ref10 doi: 10.1038/s42256-023-00714-5 – volume-title: Proc. Int. Conf. Learn. Represent. (ICLR) ident: ref42 article-title: An image is worth 16×16 words: Transformers for image recognition at scale – ident: ref35 doi: 10.1590/1980-57642016dn11-010002 – ident: ref15 doi: 10.1016/j.neuroimage.2021.118721 – volume: 1 start-page: 4171 year: 2018 ident: ref16 article-title: BERT: Pre-training of deep bidirectional transformers for language understanding publication-title: North Amer. Chapter Assoc. Comput. Linguistics – ident: ref33 doi: 10.24003/emitter.v5i1.165 – ident: ref9 doi: 10.1109/JBHI.2023.3335854 – ident: ref36 doi: 10.1109/CCMB.2013.6609178 – ident: ref5 doi: 10.1111/j.1528-1167.2006.00655.x – ident: ref14 doi: 10.1038/s41467-024-45922-8 – ident: ref50 doi: 10.1109/EMBC46164.2021.9629837 – ident: ref20 doi: 10.1109/CVPR42600.2020.00674
SSID	ssj0000605649
Score	2.4929416
Snippet	Electroencephalography (EEG) plays a crucial role in neuroscience research and clinical practice, but it remains limited by nonuniform data, noise, and...
SourceID	proquest pubmed crossref ieee
SourceType	Aggregation Database Index Database Publisher
StartPage	17664
SubjectTerms	Algorithms Autoencoders Brain - physiology Brain modeling Databases, Factual Decoding Electrodes Electroencephalography Electroencephalography (EEG) Electroencephalography - methods Feature extraction Humans Machine Learning masked autoencoder motion intention recognition Motors Neural Networks, Computer Noise Representation learning Signal Processing, Computer-Assisted signal quality enhancement Signal-To-Noise Ratio Spatio-Temporal Analysis Spatiotemporal phenomena
Title	DMAE-EEG: A Pretraining Framework for EEG Spatiotemporal Representation Learning
URI	https://ieeexplore.ieee.org/document/11062976 https://www.ncbi.nlm.nih.gov/pubmed/40601454 https://www.proquest.com/docview/3226713554
Volume	36
WOSCitedRecordID	wos001522955400001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText	1
inHoldings	1
isFullTextHit
isPrint
journalDatabaseRights	– providerCode: PRVIEE databaseName: IEEE Electronic Library (IEL) customDbUrl: eissn: 2162-2388 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0000605649 issn: 2162-237X databaseCode: RIE dateStart: 20120101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1Lb9QwEB7RCqFeWh4FlkdlJG7IbdaxY5vbCnbhAKsVLdLeIsceIySURdvd_n7GjlPEoQduiWQ70Xy25xt7HgBvQyc6FyrDo_OKE_933NngOAoRiCAbQrrLxSb0cmnWa7sqweo5FgYRs_MZnqfHfJcfNn6fjsouSFU1gvTnARxorYdgrdsDlYqIeZPprpg2gotar8cgmcpeXC2XXy7JHBTqPGX8IlZ0BA9kSkYilfxHJ-UiK3fzzax3Fif_-ccP4bgQTDYbZsQjuIf9YzgZizewspafwOrj19mcz-ef3rMZW21xrBXBFqO7FiM-y6gBu8xe1yWJ1S_2LTvPlpilnpUMrT9O4ftifvXhMy_lFbgXWu947dCjrePURE92idXKKYxRG9tIb6dBopaBCIf0sSYrMJjYYONdpaqAXads_RQO-02Pz4GZTvigamkUWmli1TWmoveO-tMOEsUE3o0Cbn8PWTTabH1Uts3ItAmZtiAzgdMkyb8tixAn8GYEpaVFkG42XI-b_XVLu1KTag0qOYFnA1q3vUeQX9wx6ks4Sh8fHPReweFuu8fXcN_f7H5eb89opq3NWZ5pfwCTtsyu
linkProvider	IEEE
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LbxMxEB6VFkEvLY8C6QOMxA253XjtXbu3qE1aRLqKaJByW3n9QEhog9KE38_Y6y3qoQduu5JtWfPZnm_seQB8sg1rtM0k9doIivxfU62spo4xiwRZItJNLDZRVpVcLNQsBavHWBjnXHQ-c6fhM77l26XZhKuyM1RVBUP9-QR2BOds2IVr3V-pZEjNi0h42bBglOXlog-TydTZvKqmt2gQMnEacn4hL9qFZzykI-GCP9BKsczK44wzap7J_n_O-QXsJYpJRt2aeAlbrn0F-335BpJ282uYXd6MxnQ8vjonIzJbub5aBJn0DlsEGS3BBuQ2-l2nNFa_yLfoPpuillqScrT-OIDvk_H84pqmAgvUsLJc01w741Tuh9IbtExUKbRw3pdSFdyooeWu5BYpBzc-RzvQSl-4wuhMZNY1jVD5G9hul617B0Q2zFiRcymc4tJnTSEz_G-wP54hng3gcy_g-neXR6OO9kem6ohMHZCpEzIDOAiS_NcyCXEAH3tQatwG4W1Dt265uavxXCpCtUHBB_C2Q-u-dw_y4SOjfoDn1_ObaT39Un09gt0wkc5d7xi216uNO4Gn5s_6593qfVxvfwGTac8N
openUrl	ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=DMAE-EEG%3A+A+Pretraining+Framework+for+EEG+Spatiotemporal+Representation+Learning&rft.jtitle=IEEE+transaction+on+neural+networks+and+learning+systems&rft.au=Zhang%2C+Yifan&rft.au=Yu%2C+Yang&rft.au=Li%2C+Hao&rft.au=Wu%2C+Anqi&rft.date=2025-10-01&rft.issn=2162-237X&rft.eissn=2162-2388&rft.volume=36&rft.issue=10&rft.spage=17664&rft.epage=17678&rft_id=info:doi/10.1109%2FTNNLS.2025.3581991&rft.externalDBID=n%2Fa&rft.externalDocID=10_1109_TNNLS_2025_3581991
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2162-237X&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2162-237X&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2162-237X&client=summon