Momentum Contrastive Voxel-wise Representation Learning for Semi-supervised Volumetric Medical Image Segmentation
Contrastive learning (CL) aims to learn useful representation without relying on expert annotations in the context of medical image segmentation. Existing approaches mainly contrast a single positive vector ( , an augmentation of the same image) against a set of negatives within the entire remainder...
Saved in:
| Published in: | Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention Vol. 13434; p. 639 |
|---|---|
| Main Authors: | , , , |
| Format: | Journal Article |
| Language: | English |
| Published: |
Germany
01.01.2022
|
| Subjects: | |
| Online Access: | Get more information |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract | Contrastive learning (CL) aims to learn useful representation without relying on expert annotations in the context of medical image segmentation. Existing approaches mainly contrast a single positive vector (
, an augmentation of the same image) against a set of negatives within the entire remainder of the batch by simply mapping all input features into the same constant vector. Despite the impressive empirical performance, those methods have the following shortcomings: (1) it remains a formidable challenge to prevent the collapsing problems to trivial solutions; and (2) we argue that not all voxels within the same image are equally positive since there exist the dissimilar anatomical structures with the same image. In this work, we present a novel
ontrastive
oxel-wise
epresentation
earning (CVRL) method to effectively learn low-level and high-level features by capturing 3D spatial context and rich anatomical information along both the feature and the batch dimensions. Specifically, we first introduce a novel CL strategy to ensure feature diversity promotion among the 3D representation dimensions. We train the framework through bi-level contrastive optimization (
, low-level and high-level) on 3D images. Experiments on two benchmark datasets and different labeled settings demonstrate the superiority of our proposed framework. More importantly, we also prove that our method inherits the benefit of hardness-aware property from the standard CL approaches. Codes will be available soon. |
|---|---|
| AbstractList | Contrastive learning (CL) aims to learn useful representation without relying on expert annotations in the context of medical image segmentation. Existing approaches mainly contrast a single positive vector (i.e., an augmentation of the same image) against a set of negatives within the entire remainder of the batch by simply mapping all input features into the same constant vector. Despite the impressive empirical performance, those methods have the following shortcomings: (1) it remains a formidable challenge to prevent the collapsing problems to trivial solutions; and (2) we argue that not all voxels within the same image are equally positive since there exist the dissimilar anatomical structures with the same image. In this work, we present a novel Contrastive Voxel-wise Representation Learning (CVRL) method to effectively learn low-level and high-level features by capturing 3D spatial context and rich anatomical information along both the feature and the batch dimensions. Specifically, we first introduce a novel CL strategy to ensure feature diversity promotion among the 3D representation dimensions. We train the framework through bi-level contrastive optimization (i.e., low-level and high-level) on 3D images. Experiments on two benchmark datasets and different labeled settings demonstrate the superiority of our proposed framework. More importantly, we also prove that our method inherits the benefit of hardness-aware property from the standard CL approaches. Codes will be available soon.Contrastive learning (CL) aims to learn useful representation without relying on expert annotations in the context of medical image segmentation. Existing approaches mainly contrast a single positive vector (i.e., an augmentation of the same image) against a set of negatives within the entire remainder of the batch by simply mapping all input features into the same constant vector. Despite the impressive empirical performance, those methods have the following shortcomings: (1) it remains a formidable challenge to prevent the collapsing problems to trivial solutions; and (2) we argue that not all voxels within the same image are equally positive since there exist the dissimilar anatomical structures with the same image. In this work, we present a novel Contrastive Voxel-wise Representation Learning (CVRL) method to effectively learn low-level and high-level features by capturing 3D spatial context and rich anatomical information along both the feature and the batch dimensions. Specifically, we first introduce a novel CL strategy to ensure feature diversity promotion among the 3D representation dimensions. We train the framework through bi-level contrastive optimization (i.e., low-level and high-level) on 3D images. Experiments on two benchmark datasets and different labeled settings demonstrate the superiority of our proposed framework. More importantly, we also prove that our method inherits the benefit of hardness-aware property from the standard CL approaches. Codes will be available soon. Contrastive learning (CL) aims to learn useful representation without relying on expert annotations in the context of medical image segmentation. Existing approaches mainly contrast a single positive vector ( , an augmentation of the same image) against a set of negatives within the entire remainder of the batch by simply mapping all input features into the same constant vector. Despite the impressive empirical performance, those methods have the following shortcomings: (1) it remains a formidable challenge to prevent the collapsing problems to trivial solutions; and (2) we argue that not all voxels within the same image are equally positive since there exist the dissimilar anatomical structures with the same image. In this work, we present a novel ontrastive oxel-wise epresentation earning (CVRL) method to effectively learn low-level and high-level features by capturing 3D spatial context and rich anatomical information along both the feature and the batch dimensions. Specifically, we first introduce a novel CL strategy to ensure feature diversity promotion among the 3D representation dimensions. We train the framework through bi-level contrastive optimization ( , low-level and high-level) on 3D images. Experiments on two benchmark datasets and different labeled settings demonstrate the superiority of our proposed framework. More importantly, we also prove that our method inherits the benefit of hardness-aware property from the standard CL approaches. Codes will be available soon. |
| Author | Zhao, Ruihan Staib, Lawrence You, Chenyu Duncan, James S |
| Author_xml | – sequence: 1 givenname: Chenyu surname: You fullname: You, Chenyu organization: Electrical Engineering, Yale University, New Haven, CT USA – sequence: 2 givenname: Ruihan surname: Zhao fullname: Zhao, Ruihan organization: Electrical and Computer Engineering, The University of Texas at Austin, TX USA – sequence: 3 givenname: Lawrence surname: Staib fullname: Staib, Lawrence organization: Biomedical Engineering, Yale University, New Haven, CT USA – sequence: 4 givenname: James S surname: Duncan fullname: Duncan, James S organization: Biomedical Engineering, Yale University, New Haven, CT USA |
| BackLink | https://www.ncbi.nlm.nih.gov/pubmed/37465615$$D View this record in MEDLINE/PubMed |
| BookMark | eNo10E9LwzAYBvAcFP9Mv4FIjl6iyZom7VGGzsGGoMNrSZO3I9AkXZJO_fYW3E7v5fc8D7zX6MwHDwjdMfrIKJVPtaxIQWjBCBOcU1I1gl2gy0JyUQpWXqH9JjjweXR4EXyOKmV7APwVfqAn3zYB_oAhQpqIyjZ4vAYVvfU73IWIP8FZksYB4mGiZor1o4McrcYbMFarHq-c2sEEd-5UcYPOO9UnuD3eGdq-vmwXb2T9vlwtntdk4JRlUhupRVcLCpzrVldgDFNQClmXbWt4DaA7LWUrCyFNR1VpJO_oBMqyrdpiPkMP_7VDDPsRUm6cTRr6XnkIY2rmVVHLaamQE70_0rF1YJohWqfib3P60vwPkb5o5A |
| ContentType | Journal Article |
| DBID | NPM 7X8 |
| DOI | 10.1007/978-3-031-16440-8_61 |
| DatabaseName | PubMed MEDLINE - Academic |
| DatabaseTitle | PubMed MEDLINE - Academic |
| DatabaseTitleList | MEDLINE - Academic PubMed |
| 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: 7X8 name: MEDLINE - Academic url: https://search.proquest.com/medline sourceTypes: Aggregation Database |
| DeliveryMethod | no_fulltext_linktorsrc |
| ExternalDocumentID | 37465615 |
| Genre | Journal Article |
| GrantInformation_xml | – fundername: NCI NIH HHS grantid: R01 CA206180 |
| GroupedDBID | NPM 7X8 |
| ID | FETCH-LOGICAL-p401t-9d7c6f960e44cbc8edd1ae56795bbd49eecfc77b7367df0a5d74f0e5655b8b32 |
| IEDL.DBID | 7X8 |
| ISICitedReferencesCount | 80 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000867306400061&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| IngestDate | Thu Jul 10 18:19:43 EDT 2025 Thu Jan 02 22:33:32 EST 2025 |
| IsDoiOpenAccess | false |
| IsOpenAccess | true |
| IsPeerReviewed | true |
| IsScholarly | true |
| Keywords | Contrastive Learning Medical Image Segmentation Semi-Supervised Learning |
| Language | English |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-p401t-9d7c6f960e44cbc8edd1ae56795bbd49eecfc77b7367df0a5d74f0e5655b8b32 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
| OpenAccessLink | https://pmc.ncbi.nlm.nih.gov/articles/PMC10352821/pdf/nihms-1912991.pdf |
| PMID | 37465615 |
| PQID | 2839740137 |
| PQPubID | 23479 |
| ParticipantIDs | proquest_miscellaneous_2839740137 pubmed_primary_37465615 |
| PublicationCentury | 2000 |
| PublicationDate | 2022-01-01 |
| PublicationDateYYYYMMDD | 2022-01-01 |
| PublicationDate_xml | – month: 01 year: 2022 text: 2022-01-01 day: 01 |
| PublicationDecade | 2020 |
| PublicationPlace | Germany |
| PublicationPlace_xml | – name: Germany |
| PublicationTitle | Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention |
| PublicationTitleAlternate | Med Image Comput Comput Assist Interv |
| PublicationYear | 2022 |
| Score | 2.3913279 |
| Snippet | Contrastive learning (CL) aims to learn useful representation without relying on expert annotations in the context of medical image segmentation. Existing... |
| SourceID | proquest pubmed |
| SourceType | Aggregation Database Index Database |
| StartPage | 639 |
| Title | Momentum Contrastive Voxel-wise Representation Learning for Semi-supervised Volumetric Medical Image Segmentation |
| URI | https://www.ncbi.nlm.nih.gov/pubmed/37465615 https://www.proquest.com/docview/2839740137 |
| Volume | 13434 |
| WOSCitedRecordID | wos000867306400061&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1JSwMxFA5qPXhxwa1uRPAa7CyZTE4iYlGwpWiR3obJVgrO0s5U_fm-zKToRRC8zCkZwnsv33t5K0JXXMmAMSGIpIoSUAEp4cwYEgvuszT0vcg0nH5iw2E8mfCRc7hVLq1yhYkNUKtCWh_5NahB3kyPYzflnNipUTa66kZorKNOAKaMTelik_hHhVwb_AfBJfAsCHskTiLvd1uy0Sn9nf-eZhdtO2sS37bs30NrOt9H84HtqlAvM2xbTy3SyiIafi0-9Rv5mFUaPzfZr67oKMeuxeoUg_2KX3Q2I9WytBBSaQXbLHrZNv7YxXTwYwYYBAun2eoXB2jcvx_fPRA3WYGUcOiacMVkZODxosNQChlrpbxU04hxKoQKudbSSGAhCyKmTC-lioWmBwsoFbEI_EO0kRe5PkbYV8YADAQS7j1gruK-jbsZMOSEiIDkXXS5ImACgmujEWmui2WVfJOwi45aLiRl22EjCZht4-bRkz_sPkVbvi1JaNwiZ6hj4Nrqc7Qp3-tZtbhoJAK-w9HgC77cxfI |
| linkProvider | ProQuest |
| 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=Momentum+Contrastive+Voxel-wise+Representation+Learning+for+Semi-supervised+Volumetric+Medical+Image+Segmentation&rft.jtitle=Medical+image+computing+and+computer-assisted+intervention+%3A+MICCAI+...+International+Conference+on+Medical+Image+Computing+and+Computer-Assisted+Intervention&rft.au=You%2C+Chenyu&rft.au=Zhao%2C+Ruihan&rft.au=Staib%2C+Lawrence&rft.au=Duncan%2C+James+S&rft.date=2022-01-01&rft.volume=13434&rft.spage=639&rft_id=info:doi/10.1007%2F978-3-031-16440-8_61&rft.externalDBID=NO_FULL_TEXT |