Mutual Evidential Deep Learning for Semi-supervised Medical Image Segmentation

Existing semi-supervised medical segmentation co-learning frameworks have realized that model performance can be diminished by the biases in model recognition caused by low-quality pseudo-labels. Due to the averaging nature of their pseudo-label integration strategy, they fail to explore the reliabi...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Proceedings (IEEE International Conference on Bioinformatics and Biomedicine) s. 2010 - 2017
Hlavní autori: He, Yuanpeng, Bi, Yali, Li, Lijian, Pun, Chi-Man, Jiao, Wenpin, Jin, Zhi
Médium: Konferenčný príspevok..
Jazyk:English
Vydavateľské údaje: IEEE 03.12.2024
Predmet:
ISSN:2156-1133
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract Existing semi-supervised medical segmentation co-learning frameworks have realized that model performance can be diminished by the biases in model recognition caused by low-quality pseudo-labels. Due to the averaging nature of their pseudo-label integration strategy, they fail to explore the reliability of pseudo-labels from different sources. In this paper, we propose a mutual evidential deep learning (MEDL) framework that offers a potentially viable solution for pseudo-label generation in semi-supervised learning from two perspectives. First, we introduce networks with different architectures to generate complementary evidence for unlabeled samples and adopt an improved class-aware evidential fusion to guide the confident synthesis of evidential predictions sourced from diverse architectural networks. Second, utilizing the uncertainty in the fused evidence, we design an asymptotic Fisher information-based evidential learning strategy. This strategy enables the model to initially focus on unlabeled samples with more reliable pseudo-labels, gradually shifting attention to samples with lower-quality pseudo-labels while avoiding over-penalization of mislabeled classes in high data uncertainty samples. Additionally, for labeled data, we continue to adopt an uncertainty-driven asymptotic learning strategy, gradually guiding the model to focus on challenging voxels. Extensive experiments on five mainstream datasets have demonstrated that MEDL achieves state-of-the-art performance.
AbstractList Existing semi-supervised medical segmentation co-learning frameworks have realized that model performance can be diminished by the biases in model recognition caused by low-quality pseudo-labels. Due to the averaging nature of their pseudo-label integration strategy, they fail to explore the reliability of pseudo-labels from different sources. In this paper, we propose a mutual evidential deep learning (MEDL) framework that offers a potentially viable solution for pseudo-label generation in semi-supervised learning from two perspectives. First, we introduce networks with different architectures to generate complementary evidence for unlabeled samples and adopt an improved class-aware evidential fusion to guide the confident synthesis of evidential predictions sourced from diverse architectural networks. Second, utilizing the uncertainty in the fused evidence, we design an asymptotic Fisher information-based evidential learning strategy. This strategy enables the model to initially focus on unlabeled samples with more reliable pseudo-labels, gradually shifting attention to samples with lower-quality pseudo-labels while avoiding over-penalization of mislabeled classes in high data uncertainty samples. Additionally, for labeled data, we continue to adopt an uncertainty-driven asymptotic learning strategy, gradually guiding the model to focus on challenging voxels. Extensive experiments on five mainstream datasets have demonstrated that MEDL achieves state-of-the-art performance.
Author He, Yuanpeng
Li, Lijian
Pun, Chi-Man
Jiao, Wenpin
Bi, Yali
Jin, Zhi
Author_xml – sequence: 1
  givenname: Yuanpeng
  surname: He
  fullname: He, Yuanpeng
  email: heyuanpeng@stu.pku.edu.cn
  organization: Peking University,Key Laboratory of High Confidence Software Technologies (MOE) School of Computer Science,Beijng,China
– sequence: 2
  givenname: Yali
  surname: Bi
  fullname: Bi, Yali
  email: biyali812@outlook.com
  organization: Southwest University,College of Computer and Information Science School of Software,Chongqing,China
– sequence: 3
  givenname: Lijian
  surname: Li
  fullname: Li, Lijian
  email: mc35305@umac.mo
  organization: Science University of Macau,Department of Computer and Information,Macau,China
– sequence: 4
  givenname: Chi-Man
  surname: Pun
  fullname: Pun, Chi-Man
  email: cmpun@um.edu.mo
  organization: Science University of Macau,Department of Computer and Information,Macau,China
– sequence: 5
  givenname: Wenpin
  surname: Jiao
  fullname: Jiao, Wenpin
  email: jwp@pku.edu.cn
  organization: Peking University,Key Laboratory of High Confidence Software Technologies (MOE) School of Computer Science,Beijng,China
– sequence: 6
  givenname: Zhi
  surname: Jin
  fullname: Jin, Zhi
  email: zhijin@pku.edu.cn
  organization: Peking University,Key Laboratory of High Confidence Software Technologies (MOE) School of Computer Science,Beijng,China
BookMark eNo1UNtKw0AUXEXBWvsHgvmB1HP2lt1HW6sGEn2w72XTPSkrTRpyKfj3LqhPMzAXhrllV-2pJcYeEJaIYB9X-arUXHC15MDlEsFwDmAu2MJm1ggFwmjO9SWbcVQ6RRTihi2G4QsAMCqIasbey2mc3DHZnIOndgyRPhN1SUGub0N7SOpTn3xSE9Jh6qg_h4F8UpIP--jMG3egqB6aGHVjOLV37Lp2x4EWfzhn25fNdv2WFh-v-fqpSANmekwFWe6MdxKkr6QAWSsLXBFRRiQM1pY0WfAg97LSWFVxq0dulamUEk7M2f1vbYiRXdeHxvXfu_8HxA99_1Gc
ContentType Conference Proceeding
DBID 6IE
6IL
CBEJK
RIE
RIL
DOI 10.1109/BIBM62325.2024.10822008
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP All) 1998-Present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Biology
EISBN 9798350386226
EISSN 2156-1133
EndPage 2017
ExternalDocumentID 10822008
Genre orig-research
GrantInformation_xml – fundername: National Natural Science Foundation of China
  funderid: 10.13039/501100001809
GroupedDBID 6IE
6IF
6IH
6IK
6IL
6IN
AAJGR
ABLEC
ADZIZ
ALMA_UNASSIGNED_HOLDINGS
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
IEGSK
IPLJI
OCL
RIE
RIL
ID FETCH-LOGICAL-i176t-3e92a8da404db4304f59025eee7ee381f9e6e90d04c4b61bb115d12958b553a3
IEDL.DBID RIE
ISICitedReferencesCount 0
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001446153502017&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
IngestDate Wed Jan 22 08:32:21 EST 2025
IsPeerReviewed false
IsScholarly false
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i176t-3e92a8da404db4304f59025eee7ee381f9e6e90d04c4b61bb115d12958b553a3
PageCount 8
ParticipantIDs ieee_primary_10822008
PublicationCentury 2000
PublicationDate 2024-Dec.-3
PublicationDateYYYYMMDD 2024-12-03
PublicationDate_xml – month: 12
  year: 2024
  text: 2024-Dec.-3
  day: 03
PublicationDecade 2020
PublicationTitle Proceedings (IEEE International Conference on Bioinformatics and Biomedicine)
PublicationTitleAbbrev BIBM
PublicationYear 2024
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssj0001622115
Score 1.9353065
Snippet Existing semi-supervised medical segmentation co-learning frameworks have realized that model performance can be diminished by the biases in model recognition...
SourceID ieee
SourceType Publisher
StartPage 2010
SubjectTerms Biomedical imaging
Class-aware evidential fusion
Data models
Deep learning
Heterogeneous networks
Image segmentation
Mutual evidential deep learning
Predictive models
Reliability
Semi-supervised medical segmentation
Semisupervised learning
Training
Uncertainty
Title Mutual Evidential Deep Learning for Semi-supervised Medical Image Segmentation
URI https://ieeexplore.ieee.org/document/10822008
WOSCitedRecordID wos001446153502017&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
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1NTwIxEG2UaOLJL4z4lR68Lu5u2217RSVygJDIgRvZtgMhkYUAa-K_d9pdJB48eGt206SZyfR12nnzCHnkjFtwwCOmck_JcSJSFkcI_cZKlnuICWITcjBQ47Ee1mT1wIUBgFB8Bm0_DG_5bmlLf1WGEY5wFqi9h1JmFVlrf6GSpZjMiLqGK4n1U6fX6SO6pwLTwJS3d7N_6agEGOme_nMBZ6S5J-TR4Q_UnJMDKC7IcaUj-XVJBv3S00BoJRGKMftBXwBWtG6eOqN4MqXvsJhHm3LlN4cNOFo_0dDeArcU_Dtb1DSkoklG3dfR81tUCyVE80Rm24iBTnPlch5zZziL-dQ3ZRG4fAmAkDzVkIGOXcwtN1liDNrJoReEMkKwnF2RRrEs4JpQI9JMcs1Tayw6USvwSVOCpxwMVcVVizS9VSarqhXGZGeQmz--35ITb_tQ_8HuSGO7LuGeHNnP7XyzfggO_AYkB5n5
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3PT8IwFG4MavTkLwz-7sHrcFvbrbuiEoiwkMiBG1nbByGRQYCZ-N_72g2JBw_emi1Nmvfy-vW173sfIY-ccQ0GuMdkZik5RnhS4wihX-mYZRZinNhEnKZyNEoGFVndcWEAwBWfQdMO3Vu-WejCXpVhhCOcOWrvvuA89Eu61u5KJQoxnRFVFVfgJ0-tbquP-B4KTARD3tzO_6Wk4oCkffLPJZyS-o6SRwc_YHNG9iA_J4elkuTXBUn7hSWC0FIkFKP2g74ALGnVPnVK8WxK32E-89bF0m4PazC0eqSh3TluKvh3Oq-ISHmdDNuvw-eOV0kleLMgjjYegyTMpMm4z43izOcT25ZF4PJjAATlSQIRJL7xueYqCpRCOxn0g5BKCJaxS1LLFzk0CFUijGKe8FArjW5MJNi0KcBzDgar5PKK1K1VxsuyGcZ4a5DrP74_kKPOsN8b97rp2w05tn5w1SDsltQ2qwLuyIH-3MzWq3vnzG9ZUJ1A
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%3Abook&rft.genre=proceeding&rft.title=Proceedings+%28IEEE+International+Conference+on+Bioinformatics+and+Biomedicine%29&rft.atitle=Mutual+Evidential+Deep+Learning+for+Semi-supervised+Medical+Image+Segmentation&rft.au=He%2C+Yuanpeng&rft.au=Bi%2C+Yali&rft.au=Li%2C+Lijian&rft.au=Pun%2C+Chi-Man&rft.date=2024-12-03&rft.pub=IEEE&rft.eissn=2156-1133&rft.spage=2010&rft.epage=2017&rft_id=info:doi/10.1109%2FBIBM62325.2024.10822008&rft.externalDocID=10822008