Semi-supervised medical image segmentation using adversarial consistency learning and dynamic convolution network

Popular semi-supervised medical image segmentation networks often suffer from error supervision from unlabeled data since they usually use consistency learning under different data perturbations to regularize model training. These networks ignore the relationship between labeled and unlabeled data,...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	IEEE transactions on medical imaging Jg. 42; H. 5; S. 1
Hauptverfasser:	Lei, Tao, Zhang, Dong, Du, Xiaogang, Wang, Xuan, Wan, Yong, Nandi, Asoke K.
Format:	Journal Article
Sprache:	Englisch
Veröffentlicht:	United States IEEE 01.05.2023 The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Schlagworte:	Adversarial learning Adversarial machine learning Computer networks Consistency Convolution Data models Dynamic convolution Image processing Image Processing, Computer-Assisted Image segmentation Learning Medical diagnosis Medical diagnostic imaging Medical image segmentation Medical imaging Networks Perturbation Perturbation methods Pixels Semi-supervised learning Spine Supervised Machine Learning Training Uncertainty
ISSN:	0278-0062, 1558-254X, 1558-254X
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Abstract	Popular semi-supervised medical image segmentation networks often suffer from error supervision from unlabeled data since they usually use consistency learning under different data perturbations to regularize model training. These networks ignore the relationship between labeled and unlabeled data, and only compute single pixel-level consistency leading to uncertain prediction results. Besides, these networks often require a large number of parameters since their backbone networks are designed depending on supervised image segmentation tasks. Moreover, these networks often face a high over-fitting risk since a small number of training samples are popular for semi-supervised image segmentation. To address the above problems, in this paper, we propose a novel adversarial self-ensembling network using dynamic convolution (ASE-Net) for semi-supervised medical image segmentation. First, we use an adversarial consistency training strategy (ACTS) that employs two discriminators based on consistency learning to obtain prior relationships between labeled and unlabeled data. The ACTS can simultaneously compute pixel-level and image-level consistency of unlabeled data under different data perturbations to improve the prediction quality of labels. Second, we design a dynamic convolution-based bidirectional attention component (DyBAC) that can be embedded in any segmentation network, aiming at adaptively adjusting the weights of ASE-Net based on the structural information of input samples. This component effectively improves the feature representation ability of ASE-Net and reduces the overfitting risk of the network. The proposed ASE-Net has been extensively tested on three publicly available datasets, and experiments indicate that ASE-Net is superior to state-of-the-art networks, and reduces computational costs and memory overhead. The code is available at: https://github.com/SUST-reynole/ASE-Net.
AbstractList	Popular semi-supervised medical image segmentation networks often suffer from error supervision from unlabeled data since they usually use consistency learning under different data perturbations to regularize model training. These networks ignore the relationship between labeled and unlabeled data, and only compute single pixel-level consistency leading to uncertain prediction results. Besides, these networks often require a large number of parameters since their backbone networks are designed depending on supervised image segmentation tasks. Moreover, these networks often face a high over-fitting risk since a small number of training samples are popular for semi-supervised image segmentation. To address the above problems, in this paper, we propose a novel adversarial self-ensembling network using dynamic convolution (ASE-Net) for semi-supervised medical image segmentation. First, we use an adversarial consistency training strategy (ACTS) that employs two discriminators based on consistency learning to obtain prior relationships between labeled and unlabeled data. The ACTS can simultaneously compute pixel-level and image-level consistency of unlabeled data under different data perturbations to improve the prediction quality of labels. Second, we design a dynamic convolution-based bidirectional attention component (DyBAC) that can be embedded in any segmentation network, aiming at adaptively adjusting the weights of ASE-Net based on the structural information of input samples. This component effectively improves the feature representation ability of ASE-Net and reduces the overfitting risk of the network. The proposed ASE-Net has been extensively tested on three publicly available datasets, and experiments indicate that ASE-Net is superior to state-of-the-art networks, and reduces computational costs and memory overhead. The code is available at: https://github.com/SUST-reynole/ASE-Net. Popular semi-supervised medical image segmentation networks often suffer from error supervision from unlabeled data since they usually use consistency learning under different data perturbations to regularize model training. These networks ignore the relationship between labeled and unlabeled data, and only compute single pixel-level consistency leading to uncertain prediction results. Besides, these networks often require a large number of parameters since their backbone networks are designed depending on supervised image segmentation tasks. Moreover, these networks often face a high over-fitting risk since a small number of training samples are popular for semi-supervised image segmentation. To address the above problems, in this paper, we propose a novel adversarial self-ensembling network using dynamic convolution (ASE-Net) for semi-supervised medical image segmentation. First, we use an adversarial consistency training strategy (ACTS) that employs two discriminators based on consistency learning to obtain prior relationships between labeled and unlabeled data. The ACTS can simultaneously compute pixel-level and image-level consistency of unlabeled data under different data perturbations to improve the prediction quality of labels. Second, we design a dynamic convolution-based bidirectional attention component (DyBAC) that can be embedded in any segmentation network, aiming at adaptively adjusting the weights of ASE-Net based on the structural information of input samples. This component effectively improves the feature representation ability of ASE-Net and reduces the overfitting risk of the network. The proposed ASE-Net has been extensively tested on three publicly available datasets, and experiments indicate that ASE-Net is superior to state-of-the-art networks, and reduces computational costs and memory overhead. The code is available at: https://github.com/SUST-reynole/ASE-Nethttps://github.com/SUST-reynole/ASE-Net . Popular semi-supervised medical image segmentation networks often suffer from error supervision from unlabeled data since they usually use consistency learning under different data perturbations to regularize model training. These networks ignore the relationship between labeled and unlabeled data, and only compute single pixel-level consistency leading to uncertain prediction results. Besides, these networks often require a large number of parameters since their backbone networks are designed depending on supervised image segmentation tasks. Moreover, these networks often face a high over-fitting risk since a small number of training samples are popular for semi-supervised image segmentation. To address the above problems, in this paper, we propose a novel adversarial self-ensembling network using dynamic convolution (ASE-Net) for semi-supervised medical image segmentation. First, we use an adversarial consistency training strategy (ACTS) that employs two discriminators based on consistency learning to obtain prior relationships between labeled and unlabeled data. The ACTS can simultaneously compute pixel-level and image-level consistency of unlabeled data under different data perturbations to improve the prediction quality of labels. Second, we design a dynamic convolution-based bidirectional attention component (DyBAC) that can be embedded in any segmentation network, aiming at adaptively adjusting the weights of ASE-Net based on the structural information of input samples. This component effectively improves the feature representation ability of ASE-Net and reduces the overfitting risk of the network. The proposed ASE-Net has been extensively tested on three publicly available datasets, and experiments indicate that ASE-Net is superior to state-of-the-art networks, and reduces computational costs and memory overhead. The code is available at: https://github.com/SUST-reynole/ASE-Nethttps://github.com/SUST-reynole/ASE-Net.Popular semi-supervised medical image segmentation networks often suffer from error supervision from unlabeled data since they usually use consistency learning under different data perturbations to regularize model training. These networks ignore the relationship between labeled and unlabeled data, and only compute single pixel-level consistency leading to uncertain prediction results. Besides, these networks often require a large number of parameters since their backbone networks are designed depending on supervised image segmentation tasks. Moreover, these networks often face a high over-fitting risk since a small number of training samples are popular for semi-supervised image segmentation. To address the above problems, in this paper, we propose a novel adversarial self-ensembling network using dynamic convolution (ASE-Net) for semi-supervised medical image segmentation. First, we use an adversarial consistency training strategy (ACTS) that employs two discriminators based on consistency learning to obtain prior relationships between labeled and unlabeled data. The ACTS can simultaneously compute pixel-level and image-level consistency of unlabeled data under different data perturbations to improve the prediction quality of labels. Second, we design a dynamic convolution-based bidirectional attention component (DyBAC) that can be embedded in any segmentation network, aiming at adaptively adjusting the weights of ASE-Net based on the structural information of input samples. This component effectively improves the feature representation ability of ASE-Net and reduces the overfitting risk of the network. The proposed ASE-Net has been extensively tested on three publicly available datasets, and experiments indicate that ASE-Net is superior to state-of-the-art networks, and reduces computational costs and memory overhead. The code is available at: https://github.com/SUST-reynole/ASE-Nethttps://github.com/SUST-reynole/ASE-Net.
Author	Lei, Tao Wang, Xuan Zhang, Dong Wan, Yong Nandi, Asoke K. Du, Xiaogang
Author_xml	– sequence: 1 givenname: Tao orcidid: 0000-0002-2104-9298 surname: Lei fullname: Lei, Tao organization: Shaanxi Joint Laboratory of Artificial Intelligence, Shaanxi University of Science and Technology, Xi'an, China – sequence: 2 givenname: Dong surname: Zhang fullname: Zhang, Dong organization: Shaanxi Joint Laboratory of Artificial Intelligence, Shaanxi University of Science and Technology, Xi'an, China – sequence: 3 givenname: Xiaogang surname: Du fullname: Du, Xiaogang organization: Shaanxi Joint Laboratory of Artificial Intelligence, Shaanxi University of Science and Technology, Xi'an, China – sequence: 4 givenname: Xuan orcidid: 0000-0002-0842-6511 surname: Wang fullname: Wang, Xuan organization: Department of Electrical and Computer Engineering, University of Wisconsin-Madison, Madison, WI, US – sequence: 5 givenname: Yong surname: Wan fullname: Wan, Yong organization: Department of Geriatric Surgery, First Affiliated Hospital, Xi'an Jiaotong University, Xi'an, China – sequence: 6 givenname: Asoke K. orcidid: 0000-0001-6248-2875 surname: Nandi fullname: Nandi, Asoke K. organization: Department of Electronic and Electrical Engineering, Brunel University London, Uxbridge, U.K
BackLink	https://www.ncbi.nlm.nih.gov/pubmed/36449588$$D View this record in MEDLINE/PubMed
BookMark	eNp9kc9LHDEYhoModbW9C4IM9NLLrEkmySRHkf4QFA8q9BayyTdLdCZZk5kt-9836649eOjpuzzPy8f7nqDDEAMgdEbwnBCsLh_vbuYUUzpvKOVCtgdoRjiXNeXs9yGaYdrKGmNBj9FJzs8YE8ax-oSOG8GY4lLO0OsDDL7O0wrS2mdw1QDOW9NXfjBLqDIsBwijGX0M1ZR9WFbGrSFlk3yBbAzZ5xGC3VQ9mBTegOAqtwlm8HYLrGM_vekBxj8xvXxGR53pM3zZ31P09OP74_Wv-vb-58311W1tG6nGumUdXXScgaOsIZR00joJxIGknQCGCcfcWQYLoSwWHBYdEQKIlY60TpKuOUXfdrmrFF8nyKMefLbQ9yZAnLKmLWtKBhe8oF8_oM9xSqF8p6nEUjZcKVyoiz01LUpLepVKR2mj38ssgNgBNsWcE3Ta-l11YzK-1wTr7Wq6rKa3q-n9akXEH8T37P8o5zvFA8A_XCkhJCPNX_drpCg
CODEN	ITMID4
CitedBy_id	crossref_primary_10_1002_ima_70091 crossref_primary_10_1109_TIP_2024_3463412 crossref_primary_10_1016_j_compmedimag_2024_102474 crossref_primary_10_1016_j_neunet_2025_107142 crossref_primary_10_1002_mp_17323 crossref_primary_10_1016_j_medengphy_2025_104337 crossref_primary_10_1016_j_patcog_2024_110856 crossref_primary_10_1007_s11227_025_07824_4 crossref_primary_10_1007_s11760_024_03536_z crossref_primary_10_1016_j_ijtst_2025_07_002 crossref_primary_10_1109_TIM_2024_3400343 crossref_primary_10_1109_TIM_2024_3488143 crossref_primary_10_3390_bioengineering11121229 crossref_primary_10_1109_TRPMS_2023_3286866 crossref_primary_10_1016_j_compmedimag_2025_102517 crossref_primary_10_1016_j_knosys_2025_113451 crossref_primary_10_1016_j_neunet_2023_11_055 crossref_primary_10_1016_j_engappai_2025_110082 crossref_primary_10_1002_ima_70163 crossref_primary_10_1016_j_eswa_2024_124816 crossref_primary_10_1109_TBME_2025_3541992 crossref_primary_10_1016_j_neucom_2025_131060 crossref_primary_10_1002_acm2_14483 crossref_primary_10_1016_j_neunet_2025_107813 crossref_primary_10_3390_info16060433 crossref_primary_10_1007_s11517_023_02946_4 crossref_primary_10_1109_TMI_2025_3563500 crossref_primary_10_1109_TMI_2025_3570310 crossref_primary_10_1016_j_asoc_2025_112904 crossref_primary_10_1016_j_neucom_2025_129782 crossref_primary_10_1007_s40747_025_01964_z crossref_primary_10_1016_j_eswa_2025_127467 crossref_primary_10_1016_j_media_2024_103450 crossref_primary_10_1016_j_bspc_2025_107561 crossref_primary_10_1016_j_compbiomed_2024_109572 crossref_primary_10_1016_j_eswa_2024_125456 crossref_primary_10_1016_j_artmed_2025_103229 crossref_primary_10_1177_08953996251367210 crossref_primary_10_1016_j_asoc_2025_113894 crossref_primary_10_1016_j_cmpb_2024_108141 crossref_primary_10_3390_bioengineering12080872 crossref_primary_10_1016_j_bspc_2025_108137 crossref_primary_10_1007_s10278_024_01122_w crossref_primary_10_1002_ima_70114 crossref_primary_10_1109_TMI_2024_3429340 crossref_primary_10_1109_TIP_2025_3536208 crossref_primary_10_3390_s25154535 crossref_primary_10_1109_TMI_2024_3468896 crossref_primary_10_1109_TMI_2025_3556310 crossref_primary_10_1016_j_engappai_2025_110528 crossref_primary_10_1109_TRPMS_2024_3473929 crossref_primary_10_1109_TMI_2024_3400840 crossref_primary_10_1016_j_bspc_2024_107450 crossref_primary_10_1016_j_compbiomed_2024_108176 crossref_primary_10_1016_j_patcog_2024_110962 crossref_primary_10_1007_s11760_024_03091_7 crossref_primary_10_1016_j_bspc_2025_107956 crossref_primary_10_1016_j_bspc_2024_107412 crossref_primary_10_1016_j_compbiomed_2024_109627 crossref_primary_10_1007_s10489_023_05158_3 crossref_primary_10_1109_TCSVT_2024_3508768 crossref_primary_10_1109_THMS_2024_3390415 crossref_primary_10_1016_j_bbe_2025_07_006 crossref_primary_10_1016_j_compbiomed_2023_107879 crossref_primary_10_1016_j_media_2025_103773 crossref_primary_10_1016_j_knosys_2025_113785 crossref_primary_10_1109_TIM_2025_3547131 crossref_primary_10_1109_TRPMS_2024_3465561 crossref_primary_10_1007_s00500_025_10572_5 crossref_primary_10_1109_TMI_2025_3556482 crossref_primary_10_3390_app13137921
Cites_doi	10.1109/CVPR42600.2020.01269 10.1109/TNNLS.2020.2995319 10.1109/TMI.2018.2845918 10.1109/TMI.2021.3117888 10.1109/CVPR46437.2021.01214 10.1016/j.media.2021.102146 10.1109/TMI.2020.3035253 10.1109/TPAMI.2021.3117837 10.1007/978-3-030-32245-8_67 10.1109/CVPR46437.2021.00264 10.1109/TMI.2019.2959609 10.1016/j.media.2020.101832 10.1109/CVPR42600.2020.00858 10.1049/ipr2.12419 10.1109/WACV51458.2022.00183 10.1007/978-3-030-78191-0_40 10.1016/j.neuroimage.2021.118568 10.1093/nsr/nwx106 10.1609/aaai.v35i10.17066 10.1007/978-3-030-32245-8_6 10.1007/978-3-030-87196-3_40 10.1109/JBHI.2021.3103646 10.1007/978-3-030-87196-3_45 10.1109/ICCV.2019.00683 10.1109/CVPR.2018.00745 10.1109/CVPR46437.2021.00658 10.1109/CVPR42600.2020.01104 10.1007/978-3-030-87196-3_28 10.1109/EMBC46164.2021.9629941 10.1109/TRPMS.2021.3059780 10.1109/ICASSP40776.2020.9053454 10.1109/ISBI48211.2021.9434135 10.1109/3DV.2016.79 10.1109/ICCV48922.2021.00347 10.1109/TMI.2022.3161829 10.1109/CVPR52688.2022.00422 10.1007/978-3-319-66179-7_47
ContentType	Journal Article
Copyright	Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2023
Copyright_xml	– notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2023
DBID	97E ESBDL RIA RIE AAYXX CITATION CGR CUY CVF ECM EIF NPM 7QF 7QO 7QQ 7SC 7SE 7SP 7SR 7TA 7TB 7U5 8BQ 8FD F28 FR3 H8D JG9 JQ2 KR7 L7M L~C L~D NAPCQ P64 7X8
DOI	10.1109/TMI.2022.3225687
DatabaseName	IEEE Xplore (IEEE) IEEE Xplore Open Access Journals IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE/IET Electronic Library CrossRef Medline MEDLINE MEDLINE (Ovid) MEDLINE MEDLINE PubMed Aluminium Industry Abstracts Biotechnology Research Abstracts Ceramic Abstracts Computer and Information Systems Abstracts Corrosion Abstracts Electronics & Communications Abstracts Engineered Materials Abstracts Materials Business File Mechanical & Transportation Engineering Abstracts Solid State and Superconductivity Abstracts METADEX Technology Research Database ANTE: Abstracts in New Technology & Engineering Engineering Research Database Aerospace Database Materials Research Database ProQuest Computer Science Collection Civil Engineering Abstracts Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Academic Computer and Information Systems Abstracts Professional Nursing & Allied Health Premium Biotechnology and BioEngineering Abstracts MEDLINE - Academic
DatabaseTitle	CrossRef MEDLINE Medline Complete MEDLINE with Full Text PubMed MEDLINE (Ovid) Materials Research Database Civil Engineering Abstracts Aluminium Industry Abstracts Technology Research Database Computer and Information Systems Abstracts – Academic Mechanical & Transportation Engineering Abstracts Electronics & Communications Abstracts ProQuest Computer Science Collection Computer and Information Systems Abstracts Ceramic Abstracts Materials Business File METADEX Biotechnology and BioEngineering Abstracts Computer and Information Systems Abstracts Professional Aerospace Database Nursing & Allied Health Premium Engineered Materials Abstracts Biotechnology Research Abstracts Solid State and Superconductivity Abstracts Engineering Research Database Corrosion Abstracts Advanced Technologies Database with Aerospace ANTE: Abstracts in New Technology & Engineering MEDLINE - Academic
DatabaseTitleList	Materials Research Database MEDLINE - Academic MEDLINE
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	Medicine Engineering
EISSN	1558-254X
EndPage	1
ExternalDocumentID	36449588 10_1109_TMI_2022_3225687 9966841
Genre	orig-research Research Support, Non-U.S. Gov't Journal Article
GrantInformation_xml	– fundername: Key Research and Development Program of Shaanxi grantid: 2022GY-436?2021ZDLGY08-07 – fundername: Shaanxi Joint Laboratory of Artificial Intelligence grantid: 2020SS-03 – fundername: National Natural Science Foundation of China grantid: 62271296, 61871259, 61861024 funderid: 10.13039/501100001809 – fundername: Natural Science Basic Research Program of Shaanxi grantid: 2021JC-47
GroupedDBID	--- -DZ -~X 0R~ 29I 4.4 5GY 5RE 6IK 97E AAJGR AARMG AASAJ AAWTH ABAZT ABQJQ ABVLG ACGFO ACGFS ACIWK ACNCT ACPRK AENEX AFRAH AGQYO AHBIQ AKJIK AKQYR ALMA_UNASSIGNED_HOLDINGS ASUFR ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ CS3 DU5 EBS ESBDL F5P HZ~ IFIPE IPLJI JAVBF LAI M43 MS~ O9- OCL P2P PQQKQ RIA RIE RNS RXW TAE TN5 .GJ 53G 5VS AAYXX AETIX AGSQL AI. AIBXA ALLEH CITATION EJD H~9 IBMZZ ICLAB IFJZH VH1 AAYOK CGR CUY CVF ECM EIF NPM RIG 7QF 7QO 7QQ 7SC 7SE 7SP 7SR 7TA 7TB 7U5 8BQ 8FD F28 FR3 H8D JG9 JQ2 KR7 L7M L~C L~D NAPCQ P64 7X8
ID	FETCH-LOGICAL-c389t-74f2bf54ed243121f8cd8e1de82f6e401505dc4eb69c065ebf166e1c8d17d81f3
IEDL.DBID	RIE
ISICitedReferencesCount	107
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000982483400003&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	0278-0062 1558-254X
IngestDate	Sun Sep 28 08:50:14 EDT 2025 Sun Nov 30 03:50:58 EST 2025 Thu Apr 03 07:03:17 EDT 2025 Sat Nov 29 05:14:10 EST 2025 Tue Nov 18 22:11:23 EST 2025 Wed Aug 27 02:29:10 EDT 2025
IsDoiOpenAccess	true
IsOpenAccess	true
IsPeerReviewed	false
IsScholarly	true
Issue	5
Language	English
License	https://creativecommons.org/licenses/by-nc-nd/4.0
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c389t-74f2bf54ed243121f8cd8e1de82f6e401505dc4eb69c065ebf166e1c8d17d81f3
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 content type line 23
ORCID	0000-0002-0842-6511 0000-0001-6248-2875 0000-0002-2104-9298 0000-0002-0612-6064
OpenAccessLink	https://ieeexplore.ieee.org/document/9966841
PMID	36449588
PQID	2808835990
PQPubID	85460
PageCount	1
ParticipantIDs	crossref_citationtrail_10_1109_TMI_2022_3225687 proquest_miscellaneous_2743505565 proquest_journals_2808835990 ieee_primary_9966841 pubmed_primary_36449588 crossref_primary_10_1109_TMI_2022_3225687
PublicationCentury	2000
PublicationDate	2023-05-01
PublicationDateYYYYMMDD	2023-05-01
PublicationDate_xml	– month: 05 year: 2023 text: 2023-05-01 day: 01
PublicationDecade	2020
PublicationPlace	United States
PublicationPlace_xml	– name: United States – name: New York
PublicationTitle	IEEE transactions on medical imaging
PublicationTitleAbbrev	TMI
PublicationTitleAlternate	IEEE Trans Med Imaging
PublicationYear	2023
Publisher	IEEE The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Publisher_xml	– name: IEEE – name: The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
References	ref13 ref12 feng (ref15) 2020 hung (ref11) 2018 ref10 ref16 ref19 ref18 samuli (ref27) 2017; 4 ref51 ref50 ref46 kozinski (ref28) 2017 ref47 ref42 ref44 ref43 sohn (ref7) 2020; 33 yang (ref41) 2019; 32 zhu (ref14) 2020 ref8 ref9 ref4 ref3 ref6 wang (ref24) 2021 ref40 ref35 ref34 ref37 ref36 ref31 ref30 ref33 ref32 ref2 ref39 ref38 codella (ref49) 2019 ref23 ref26 ref25 ref20 ref22 ref21 xiang (ref17) 2021 ronneberger (ref1) 2015 ref29 bilic (ref48) 2019 li (ref45) 2021 tarvainen (ref5) 2017
References_xml	– ident: ref10 doi: 10.1109/CVPR42600.2020.01269 – ident: ref6 doi: 10.1109/TNNLS.2020.2995319 – volume: 33 start-page: 596 year: 2020 ident: ref7 article-title: FixMatch: Simplifying semi-supervised learning with consistency and confidence publication-title: Proc Adv Neural Inf Process Syst – ident: ref3 doi: 10.1109/TMI.2018.2845918 – ident: ref16 doi: 10.1109/TMI.2021.3117888 – ident: ref43 doi: 10.1109/CVPR46437.2021.01214 – year: 2019 ident: ref48 article-title: The liver tumor segmentation benchmark (LiTS) publication-title: arXiv 1901 04056 – year: 2021 ident: ref17 article-title: Self-ensembling contrastive learning for semi-supervised medical image segmentation publication-title: arXiv 2105 12924 – ident: ref21 doi: 10.1016/j.media.2021.102146 – ident: ref42 doi: 10.1109/TMI.2020.3035253 – start-page: 1 year: 2021 ident: ref45 article-title: Omni-dimensional dynamic convolution publication-title: Proc Int Conf Learn Represent – ident: ref38 doi: 10.1109/TPAMI.2021.3117837 – ident: ref9 doi: 10.1007/978-3-030-32245-8_67 – ident: ref8 doi: 10.1109/CVPR46437.2021.00264 – ident: ref2 doi: 10.1109/TMI.2019.2959609 – volume: 32 start-page: 1 year: 2019 ident: ref41 article-title: CondConv: Conditionally parameterized convolutions for efficient inference publication-title: Proc Adv Neural Inf Process Syst – ident: ref50 doi: 10.1016/j.media.2020.101832 – ident: ref39 doi: 10.1109/CVPR42600.2020.00858 – ident: ref22 doi: 10.1049/ipr2.12419 – start-page: 1 year: 2017 ident: ref28 article-title: An adversarial regularisation for semi-supervised training of structured output neural networks publication-title: Proc Conf Neural Inf Process Syst – start-page: 206 year: 2021 ident: ref24 article-title: Boundary-aware transformers for skin lesion segmentation publication-title: Proc Int Conf Med Image Comput Comput -Assist Intervent – ident: ref35 doi: 10.1109/WACV51458.2022.00183 – ident: ref29 doi: 10.1007/978-3-030-78191-0_40 – ident: ref13 doi: 10.1016/j.neuroimage.2021.118568 – ident: ref4 doi: 10.1093/nsr/nwx106 – ident: ref33 doi: 10.1609/aaai.v35i10.17066 – year: 2020 ident: ref15 article-title: DMT: Dynamic mutual training for semi-supervised learning publication-title: arXiv 2004 08514 – ident: ref51 doi: 10.1007/978-3-030-32245-8_6 – ident: ref34 doi: 10.1007/978-3-030-87196-3_40 – start-page: 1195 year: 2017 ident: ref5 article-title: Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results publication-title: Proc 31st Int Conf Neural Inf Process Syst – start-page: 234 year: 2015 ident: ref1 article-title: U-Net: Convolutional networks for biomedical image segmentation publication-title: Proc Int Conf Med Image Comput Comput -Assist Intervent – ident: ref20 doi: 10.1109/JBHI.2021.3103646 – year: 2019 ident: ref49 article-title: Skin lesion analysis toward melanoma detection 2018: A challenge hosted by the international skin imaging collaboration (ISIC) publication-title: arXiv 1902 03368 – ident: ref18 doi: 10.1007/978-3-030-87196-3_45 – year: 2020 ident: ref14 article-title: Improving semantic segmentation via self-training publication-title: arXiv 2004 14960 – ident: ref46 doi: 10.1109/ICCV.2019.00683 – ident: ref47 doi: 10.1109/CVPR.2018.00745 – ident: ref44 doi: 10.1109/CVPR46437.2021.00658 – ident: ref40 doi: 10.1109/CVPR42600.2020.01104 – ident: ref31 doi: 10.1007/978-3-030-87196-3_28 – volume: 4 start-page: 6 year: 2017 ident: ref27 article-title: Temporal ensembling for semi-supervised learning publication-title: Proc Int Conf Learn Represent (ICLR) – start-page: 1 year: 2018 ident: ref11 article-title: Adversarial learning for semi-supervised semantic segmentation publication-title: Proc 29th Brit Mach Vis Conf (BMVC) – ident: ref30 doi: 10.1109/EMBC46164.2021.9629941 – ident: ref23 doi: 10.1109/TRPMS.2021.3059780 – ident: ref26 doi: 10.1109/ICASSP40776.2020.9053454 – ident: ref36 doi: 10.1109/ISBI48211.2021.9434135 – ident: ref25 doi: 10.1109/3DV.2016.79 – ident: ref37 doi: 10.1109/ICCV48922.2021.00347 – ident: ref19 doi: 10.1109/TMI.2022.3161829 – ident: ref32 doi: 10.1109/CVPR52688.2022.00422 – ident: ref12 doi: 10.1007/978-3-319-66179-7_47
SSID	ssj0014509
Score	2.6594756
Snippet	Popular semi-supervised medical image segmentation networks often suffer from error supervision from unlabeled data since they usually use consistency learning...
SourceID	proquest pubmed crossref ieee
SourceType	Aggregation Database Index Database Enrichment Source Publisher
StartPage	1
SubjectTerms	Adversarial learning Adversarial machine learning Computer networks Consistency Convolution Data models Dynamic convolution Image processing Image Processing, Computer-Assisted Image segmentation Learning Medical diagnosis Medical diagnostic imaging Medical image segmentation Medical imaging Networks Perturbation Perturbation methods Pixels Semi-supervised learning Spine Supervised Machine Learning Training Uncertainty
Title	Semi-supervised medical image segmentation using adversarial consistency learning and dynamic convolution network
URI	https://ieeexplore.ieee.org/document/9966841 https://www.ncbi.nlm.nih.gov/pubmed/36449588 https://www.proquest.com/docview/2808835990 https://www.proquest.com/docview/2743505565
Volume	42
WOSCitedRecordID	wos000982483400003&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: 1558-254X dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0014509 issn: 0278-0062 databaseCode: RIE dateStart: 19820101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1baxQxFD7UIqIPXlovo7VE8EVwupvMJcmjiEUfWgQr7Nswk5yUBXe2dnYL_nvPmWRHH1TwbSBnLnC-JF_mXD6A10UgYt8GlVvv5nlZuDpvZaVz7lItaaTUdRSb0OfnZrGwn_fg7VQLg4hj8hme8OUYy_drt-VfZTPm5oar1G9prWOt1hQxKKuYzqG4Y-y8VruQ5NzOLs4-0UFQqRMGb21Yd68gGmCrUW7l1240yqv8nWmOO87pg__71odwPzFL8S5C4RHsYX8A937rN3gAd85SJP0Qvn_B1TIftle8VgzoxSpGbMRyRSuMGPBylaqSesG58ZeiZenmoWXACsdptQPT7R8i6U6QQe-FjwL3bHCTQC36mGn-GL6efrh4_zFP8gu5IxazyXUZVBeqEr0ilqFkMM4blB6NCjWW_Kuk8q7ErraOiAx2QdY1Sme81N7IUDyB_X7d4zMQqiOa46W13lg6QQai9VLLVhOZDFZ2KoPZzg2NS73JWSLjWzOeUea2IR827MMm-TCDN9MdV7Evxz9sD9k_k11yTQZHO083aeIOjTK07BYV7dEZvJqGacpxHKXtcb0lG2JdFTchqjJ4GhEyPXsHrOd_fucLuMt69TFj8gj2N9dbfAm33c1mOVwfE64X5njE9U8LXPJ3
linkProvider	IEEE
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1ba9RAFD6UVrw8eGm1plYdwRfBdHcmk2TmUcTSYncRXKFvIZlLWXCzbbNb6L_vOZnZ6IMKvgXm5ALnm5lvci4fwPvMI7GvvUi1NeNUZqZIa56XKXWp5jgiyyKITZTTqTo_19-24ONQC-Oc65PP3BFd9rF8uzRr-lU2Im6uqEp9J5dS8FCtNcQMZB4SOgT1jB0XYhOUHOvRbHKKR0Ehjgi-hSLlvQyJgM57wZVf-1EvsPJ3rtnvOcdP_u9rn8LjyC3ZpwCGZ7Dl2l149FvHwV24P4mx9D24-u4W87RbX9Jq0TnLFiFmw-YLXGNY5y4WsS6pZZQdf8FqEm_uaoIsM5RY2xHhvmVReQINWstskLgng5sIa9aGXPPn8OP4y-zzSRoFGFKDPGaVltKLxufSWYE8Q3CvjFWOW6eEL5yknyW5NdI1hTZIZVzjeVE4bpTlpVXcZy9gu1227iUw0SDRsVxrqzSeIT0Se17yukQ66TVvRAKjjRsqE7uTk0jGz6o_pYx1hT6syIdV9GECH4Y7LkNnjn_Y7pF_BrvomgQON56u4tTtKqFw4c1y3KUTeDcM46SjSErduuUabZB35dSGKE9gPyBkePYGWAd_fudbeHAym5xVZ6fTr6_gIanXh_zJQ9heXa_da7hnblbz7vpNj-47EmD01g
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=Semi-Supervised+Medical+Image+Segmentation+Using+Adversarial+Consistency+Learning+and+Dynamic+Convolution+Network&rft.jtitle=IEEE+transactions+on+medical+imaging&rft.au=Lei%2C+Tao&rft.au=Zhang%2C+Dong&rft.au=Du%2C+Xiaogang&rft.au=Wang%2C+Xuan&rft.date=2023-05-01&rft.issn=1558-254X&rft.eissn=1558-254X&rft.volume=42&rft.issue=5&rft.spage=1265&rft_id=info:doi/10.1109%2FTMI.2022.3225687&rft.externalDBID=NO_FULL_TEXT
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0278-0062&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0278-0062&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0278-0062&client=summon