Efficient few-shot medical image segmentation via self-supervised variational autoencoder

Few-shot medical image segmentation typically uses a joint model for registration and segmentation. The registration model aligns a labeled atlas with unlabeled images to form initial masks, which are then refined by the segmentation model. However, inevitable spatial misalignments during registrati...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Medical image analysis Ročník 104; s. 103637
Hlavní autoři: Zhou, Yanjie, Zhou, Feng, Xi, Fengjun, Liu, Yong, Peng, Yun, Carlson, David E., Tu, Liyun
Médium: Journal Article
Jazyk:angličtina
Vydáno: Netherlands Elsevier B.V 01.08.2025
Témata:
ISSN:1361-8415, 1361-8423, 1361-8423
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract Few-shot medical image segmentation typically uses a joint model for registration and segmentation. The registration model aligns a labeled atlas with unlabeled images to form initial masks, which are then refined by the segmentation model. However, inevitable spatial misalignments during registration can lead to inaccuracies and diminished segmentation quality. To address this, we developed EFS-MedSeg, an end-to-end model using two labeled atlases and few unlabeled images, enhanced by data augmentation and self-supervised learning. Initially, EFS-MedSeg applies a 3D random regional switch strategy to augment atlases, thereby enhancing supervision in segmentation tasks. This not only introduces variability to the training data but also enhances the model’s ability to generalize and prevents overfitting, resulting in natural and smooth label boundaries. Following this, we use a variational autoencoder for a weighted reconstruction task, focusing the model’s attention on areas with lower Dice scores to ensure accurate segmentation that conforms to the atlas image’s shape and structural appearance. Moreover, we introduce a self-contrastive module aimed at improving feature extraction, guided by anatomical structure priors, thus enhancing the model’s convergence and segmentation accuracy. Results on multi-modal medical image datasets show that EFS-MedSeg achieves performance comparable to fully-supervised methods. Moreover, it consistently surpasses the second-best method in Dice score by 1.4%, 9.1%, and 1.1% on the OASIS, BCV, and BCH datasets, respectively, highlighting its robustness and adaptability across diverse datasets. The source code will be made publicly available at: https://github.com/NoviceFodder/EFS-MedSeg. •Registration-free model enhances few-shot segmentation with 3D random regional switch.•Self-supervised framework refines low-Dice areas for precise segmentation.•Adaptive attention mechanism balances tissue volume discrepancies and boosts accuracy.•Self-contrastive module leverages anatomical priors for better features and outcomes.
AbstractList Few-shot medical image segmentation typically uses a joint model for registration and segmentation. The registration model aligns a labeled atlas with unlabeled images to form initial masks, which are then refined by the segmentation model. However, inevitable spatial misalignments during registration can lead to inaccuracies and diminished segmentation quality. To address this, we developed EFS-MedSeg, an end-to-end model using two labeled atlases and few unlabeled images, enhanced by data augmentation and self-supervised learning. Initially, EFS-MedSeg applies a 3D random regional switch strategy to augment atlases, thereby enhancing supervision in segmentation tasks. This not only introduces variability to the training data but also enhances the model's ability to generalize and prevents overfitting, resulting in natural and smooth label boundaries. Following this, we use a variational autoencoder for a weighted reconstruction task, focusing the model's attention on areas with lower Dice scores to ensure accurate segmentation that conforms to the atlas image's shape and structural appearance. Moreover, we introduce a self-contrastive module aimed at improving feature extraction, guided by anatomical structure priors, thus enhancing the model's convergence and segmentation accuracy. Results on multi-modal medical image datasets show that EFS-MedSeg achieves performance comparable to fully-supervised methods. Moreover, it consistently surpasses the second-best method in Dice score by 1.4%, 9.1%, and 1.1% on the OASIS, BCV, and BCH datasets, respectively, highlighting its robustness and adaptability across diverse datasets. The source code will be made publicly available at: https://github.com/NoviceFodder/EFS-MedSeg.Few-shot medical image segmentation typically uses a joint model for registration and segmentation. The registration model aligns a labeled atlas with unlabeled images to form initial masks, which are then refined by the segmentation model. However, inevitable spatial misalignments during registration can lead to inaccuracies and diminished segmentation quality. To address this, we developed EFS-MedSeg, an end-to-end model using two labeled atlases and few unlabeled images, enhanced by data augmentation and self-supervised learning. Initially, EFS-MedSeg applies a 3D random regional switch strategy to augment atlases, thereby enhancing supervision in segmentation tasks. This not only introduces variability to the training data but also enhances the model's ability to generalize and prevents overfitting, resulting in natural and smooth label boundaries. Following this, we use a variational autoencoder for a weighted reconstruction task, focusing the model's attention on areas with lower Dice scores to ensure accurate segmentation that conforms to the atlas image's shape and structural appearance. Moreover, we introduce a self-contrastive module aimed at improving feature extraction, guided by anatomical structure priors, thus enhancing the model's convergence and segmentation accuracy. Results on multi-modal medical image datasets show that EFS-MedSeg achieves performance comparable to fully-supervised methods. Moreover, it consistently surpasses the second-best method in Dice score by 1.4%, 9.1%, and 1.1% on the OASIS, BCV, and BCH datasets, respectively, highlighting its robustness and adaptability across diverse datasets. The source code will be made publicly available at: https://github.com/NoviceFodder/EFS-MedSeg.
Few-shot medical image segmentation typically uses a joint model for registration and segmentation. The registration model aligns a labeled atlas with unlabeled images to form initial masks, which are then refined by the segmentation model. However, inevitable spatial misalignments during registration can lead to inaccuracies and diminished segmentation quality. To address this, we developed EFS-MedSeg, an end-to-end model using two labeled atlases and few unlabeled images, enhanced by data augmentation and self-supervised learning. Initially, EFS-MedSeg applies a 3D random regional switch strategy to augment atlases, thereby enhancing supervision in segmentation tasks. This not only introduces variability to the training data but also enhances the model’s ability to generalize and prevents overfitting, resulting in natural and smooth label boundaries. Following this, we use a variational autoencoder for a weighted reconstruction task, focusing the model’s attention on areas with lower Dice scores to ensure accurate segmentation that conforms to the atlas image’s shape and structural appearance. Moreover, we introduce a self-contrastive module aimed at improving feature extraction, guided by anatomical structure priors, thus enhancing the model’s convergence and segmentation accuracy. Results on multi-modal medical image datasets show that EFS-MedSeg achieves performance comparable to fully-supervised methods. Moreover, it consistently surpasses the second-best method in Dice score by 1.4%, 9.1%, and 1.1% on the OASIS, BCV, and BCH datasets, respectively, highlighting its robustness and adaptability across diverse datasets. The source code will be made publicly available at: https://github.com/NoviceFodder/EFS-MedSeg. •Registration-free model enhances few-shot segmentation with 3D random regional switch.•Self-supervised framework refines low-Dice areas for precise segmentation.•Adaptive attention mechanism balances tissue volume discrepancies and boosts accuracy.•Self-contrastive module leverages anatomical priors for better features and outcomes.
Few-shot medical image segmentation typically uses a joint model for registration and segmentation. The registration model aligns a labeled atlas with unlabeled images to form initial masks, which are then refined by the segmentation model. However, inevitable spatial misalignments during registration can lead to inaccuracies and diminished segmentation quality. To address this, we developed EFS-MedSeg, an end-to-end model using two labeled atlases and few unlabeled images, enhanced by data augmentation and self-supervised learning. Initially, EFS-MedSeg applies a 3D random regional switch strategy to augment atlases, thereby enhancing supervision in segmentation tasks. This not only introduces variability to the training data but also enhances the model's ability to generalize and prevents overfitting, resulting in natural and smooth label boundaries. Following this, we use a variational autoencoder for a weighted reconstruction task, focusing the model's attention on areas with lower Dice scores to ensure accurate segmentation that conforms to the atlas image's shape and structural appearance. Moreover, we introduce a self-contrastive module aimed at improving feature extraction, guided by anatomical structure priors, thus enhancing the model's convergence and segmentation accuracy. Results on multi-modal medical image datasets show that EFS-MedSeg achieves performance comparable to fully-supervised methods. Moreover, it consistently surpasses the second-best method in Dice score by 1.4%, 9.1%, and 1.1% on the OASIS, BCV, and BCH datasets, respectively, highlighting its robustness and adaptability across diverse datasets. The source code will be made publicly available at: https://github.com/NoviceFodder/EFS-MedSeg.
ArticleNumber 103637
Author Carlson, David E.
Zhou, Feng
Peng, Yun
Xi, Fengjun
Zhou, Yanjie
Tu, Liyun
Liu, Yong
Author_xml – sequence: 1
  givenname: Yanjie
  surname: Zhou
  fullname: Zhou, Yanjie
  organization: School of Artificial Intelligence, Beijing University of Posts and Telecommunications, Beijing, 100876, China
– sequence: 2
  givenname: Feng
  surname: Zhou
  fullname: Zhou, Feng
  organization: School of Artificial Intelligence, Beijing University of Posts and Telecommunications, Beijing, 100876, China
– sequence: 3
  givenname: Fengjun
  surname: Xi
  fullname: Xi, Fengjun
  organization: Department of Radiology, Beijing Tiantan Hospital, Capital Medical University, Beijing 100070, China
– sequence: 4
  givenname: Yong
  surname: Liu
  fullname: Liu, Yong
  organization: School of Artificial Intelligence, Beijing University of Posts and Telecommunications, Beijing, 100876, China
– sequence: 5
  givenname: Yun
  surname: Peng
  fullname: Peng, Yun
  email: ppengyun@hotmail.com
  organization: Department of Radiology, MOE Key Laboratory of Major Diseases in Children, Beijing Children’s Hospital, Capital Medical University, National Center for Children’s Health, Beijing 100045, China
– sequence: 6
  givenname: David E.
  orcidid: 0000-0003-1005-6385
  surname: Carlson
  fullname: Carlson, David E.
  organization: Department of Civil and Environmental Engineering, and the Department of Electrical and Computer Engineering, Duke University, Durham, NC 27708, USA
– sequence: 7
  givenname: Liyun
  orcidid: 0000-0002-3389-400X
  surname: Tu
  fullname: Tu, Liyun
  email: tuliyun@bupt.edu.cn
  organization: School of Artificial Intelligence, Beijing University of Posts and Telecommunications, Beijing, 100876, China
BackLink https://www.ncbi.nlm.nih.gov/pubmed/40449308$$D View this record in MEDLINE/PubMed
BookMark eNp9kLtOwzAUhi1URC_wBEgoI0uKb7kNDKgqF6kSCwxMluscF1dpXOwkiLfHaUpHJlvnfP-R_m-KRrWtAaFrgucEk_RuO99BaeScYpqECUtZdoYmhKUkzjllo9OfJGM09X6LMc44xxdozDHnBcP5BH0stTbKQN1EGr5j_2mbqD-rZBWZndxA5GGzC2vZGFtHnZFhUOnYt3twnfFQRp105rANEdk2FmplS3CX6FzLysPV8Z2h98fl2-I5Xr0-vSweVrFihDRxkQPlOSlKBQSyggInOE2SIk1Kragsc55j0OtcalJwxQrJcKbWJKFSZzxNCzZDt8PdvbNfLfhG7IxXUFWyBtt6wSjhjKbBWEBvjmi7DiXF3oWK7kf86QgAGwDlrPcO9AkhWPTSxVYcpIteuhikh9T9kIJQszPghO-NqgA6UI0orfk3_wualosG
Cites_doi 10.1109/TMI.2019.2897538
10.1109/TMI.2022.3213983
10.1007/s11263-023-01970-z
10.1109/TAI.2023.3298303
10.1016/j.media.2022.102615
10.1109/JBHI.2019.2951024
10.1038/s41596-023-00806-x
10.1016/j.media.2021.102166
10.1109/TSP.2022.3230329
10.1038/s41551-024-01337-w
10.1109/ICCV51070.2023.01960
10.1109/CVPR52733.2024.01114
10.1109/TMI.2022.3154934
10.1162/jocn.2007.19.9.1498
10.1038/s41467-022-30695-9
10.1609/aaai.v35i2.16212
10.1038/s41592-020-01008-z
10.1109/TMI.2024.3358295
10.1186/s40537-023-00851-z
10.1109/TMI.2022.3170879
10.1016/j.media.2020.101894
10.1016/j.media.2007.06.004
10.1016/j.media.2022.102379
10.1038/s41467-024-44824-z
ContentType Journal Article
Copyright 2025 Elsevier B.V.
Copyright © 2025 Elsevier B.V. All rights reserved.
Copyright_xml – notice: 2025 Elsevier B.V.
– notice: Copyright © 2025 Elsevier B.V. All rights reserved.
DBID AAYXX
CITATION
CGR
CUY
CVF
ECM
EIF
NPM
7X8
DOI 10.1016/j.media.2025.103637
DatabaseName 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 - 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: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Medicine
Engineering
EISSN 1361-8423
ExternalDocumentID 40449308
10_1016_j_media_2025_103637
S1361841525001847
Genre Journal Article
GroupedDBID ---
--K
--M
.~1
0R~
1B1
1~.
1~5
29M
4.4
457
4G.
53G
5GY
5VS
7-5
71M
8P~
AAEDT
AAEDW
AAIKJ
AAKOC
AALRI
AAOAW
AAQFI
AAQXK
AATTM
AAXKI
AAXUO
AAYFN
AAYWO
ABBOA
ABBQC
ABJNI
ABMAC
ABMZM
ABWVN
ABXDB
ACDAQ
ACGFS
ACIEU
ACIUM
ACIWK
ACNNM
ACPRK
ACRLP
ACRPL
ACVFH
ACZNC
ADBBV
ADCNI
ADEZE
ADJOM
ADMUD
ADNMO
ADTZH
ADVLN
AEBSH
AECPX
AEIPS
AEKER
AENEX
AEUPX
AFJKZ
AFPUW
AFRAH
AFTJW
AFXIZ
AGCQF
AGHFR
AGQPQ
AGRNS
AGUBO
AGYEJ
AHJVU
AHZHX
AIALX
AIEXJ
AIGII
AIIUN
AIKHN
AITUG
AJRQY
AKBMS
AKRWK
AKYEP
ALMA_UNASSIGNED_HOLDINGS
AMRAJ
ANKPU
ANZVX
AOUOD
APXCP
ASPBG
AVWKF
AXJTR
AZFZN
BJAXD
BKOJK
BLXMC
BNPGV
C45
CAG
COF
CS3
DU5
EBS
EFJIC
EJD
EO8
EO9
EP2
EP3
F5P
FDB
FEDTE
FGOYB
FIRID
FNPLU
FYGXN
G-Q
GBLVA
GBOLZ
HVGLF
HX~
HZ~
IHE
J1W
JJJVA
KOM
M41
MO0
N9A
O-L
O9-
OAUVE
OVD
OZT
P-8
P-9
P2P
PC.
Q38
R2-
RIG
ROL
RPZ
SDF
SDG
SDP
SEL
SES
SEW
SPC
SPCBC
SSH
SST
SSV
SSZ
T5K
TEORI
UHS
~G-
9DU
AAYXX
ACLOT
CITATION
EFKBS
EFLBG
~HD
CGR
CUY
CVF
ECM
EIF
NPM
7X8
ID FETCH-LOGICAL-c311t-98e24819dce1e792e410655965dfc2ad8480efb8af194c39a307cb152af746693
ISICitedReferencesCount 0
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001502478000002&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1361-8415
1361-8423
IngestDate Thu Oct 02 22:41:30 EDT 2025
Mon Jul 21 06:03:29 EDT 2025
Sat Nov 29 07:50:29 EST 2025
Sat Jul 05 17:11:45 EDT 2025
IsPeerReviewed true
IsScholarly true
Keywords Variational autoencoder
Few-shot learning
Image reconstruction
Medical image segmentation
Language English
License Copyright © 2025 Elsevier B.V. All rights reserved.
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c311t-98e24819dce1e792e410655965dfc2ad8480efb8af194c39a307cb152af746693
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ORCID 0000-0002-3389-400X
0000-0003-1005-6385
PMID 40449308
PQID 3214326016
PQPubID 23479
ParticipantIDs proquest_miscellaneous_3214326016
pubmed_primary_40449308
crossref_primary_10_1016_j_media_2025_103637
elsevier_sciencedirect_doi_10_1016_j_media_2025_103637
PublicationCentury 2000
PublicationDate 2025-08-01
PublicationDateYYYYMMDD 2025-08-01
PublicationDate_xml – month: 08
  year: 2025
  text: 2025-08-01
  day: 01
PublicationDecade 2020
PublicationPlace Netherlands
PublicationPlace_xml – name: Netherlands
PublicationTitle Medical image analysis
PublicationTitleAlternate Med Image Anal
PublicationYear 2025
Publisher Elsevier B.V
Publisher_xml – name: Elsevier B.V
References Wang, Zhang, Kan, Shan, Chen (b56) 2020
Han, Jiang, Liu, Hu (b16) 2022
Snell, Swersky, Zemel (b41) 2017; 30
Isensee, Jaeger, Kohl, Petersen, Maier-Hein (b22) 2021; 18
Yun, Han, Oh, Chun, Choe, Yoo (b58) 2019
Tomczak, Welling (b46) 2018
Talbot, Dunson, Dzirasa, Carlson (b43) 2020
Butoi, V.I., Ortiz, J.J.G., Ma, T., Sabuncu, M.R., Guttag, J., Dalca, A.V., 2023. Universeg: Universal medical image segmentation. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 21438–21451.
Zhao, Lau, Luo, Eric, Chang, Xu (b61) 2019; 24
Ma, He, Li, Han, You, Wang (b32) 2024; 15
Zhao, Song, Ermon (b62) 2017
Fan, X., Wang, X., Gao, J., Wang, J., Luo, Z., Liu, R., 2024. Bi-level Learning of Task-Specific Decoders for Joint Registration and One-Shot Medical Image Segmentation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 11726–11735.
Avants, Epstein, Grossman, Gee (b3) 2008; 12
Liu, Aviles-Rivero, Schönlieb (b28) 2023
Avants, Tustison, Song (b4) 2009; 2
Joy, Schmon, Torr, N, Rainforth (b24) 2021
Chen, Frey, He, Segars, Li, Du (b8) 2022; 82
Oord, Li, Vinyals (b36) 2018
Tu, Talbot, Gallagher, Carlson (b47) 2022; 70
Balakrishnan, Zhao, Sabuncu, Guttag, Dalca (b5) 2019; 38
Rezende, Mohamed, Wierstra (b38) 2014
Vahdat, Kautz (b48) 2020; Vol. 33
Kang, Hu, Huang, Scott, Reyes (b25) 2022; 78
Pu, Gan, Henao, Yuan, Li, Stevens, Carin (b37) 2016
Wang, Liew, Zou, Zhou, Feng (b53) 2019
Ahn, Yoo, Sohn (b1) 2024; 132
Zhong, Zheng, Kang, Li, Yang (b64) 2017
Sekuboyina, Husseini, Bayat, Löffler, Liebl, Li, Tetteh, Kukačka, Payer, Štern (b40) 2021; 73
Kingma, Welling (b27) 2014
Wang, Cao, Wei, Wang, Ma, Wang, Meng, Zheng (b51) 2020
Lv, Wang, Shi, Zhang, Wang, Wang, Li (b31) 2022; 41
Wang, Guo, Ye, Deng, Cheng, Li, Chen, Su, Huang, Shen, Fu, Zhang, He, Qiao (b52) 2023
Madry, Makelov, Schmidt, Tsipras, Vladu (b33) 2018
Marcus, Wang, Parker, Csernansky, Morris, Buckner (b34) 2007; 19
Kingma, Rezende, Mohamed, Welling (b26) 2014; Vol. 27
Cheng, Wang, Xin, Zhou, Zhang, Shao (b9) 2024; 43
Hering, Hansen, Mok, Chung, Siebert, Häger, Lange, Kuckertz, Heldmann, Shao (b20) 2022; 42
Jiang, Veeraraghavan (b23) 2022; 41
Antonelli, Reinke, Bakas, Farahani, Kopp-Schneider, Landman, Litjens, Menze, Ronneberger, Summers (b2) 2022; 13
Billion Polak, Prusa, Khoshgoftaar (b6) 2024; 11
Zhao, Balakrishnan, Durand, Guttag, Dalca (b60) 2019
Grill, Strub, Altché, Tallec, Richemond, Buchatskaya, Doersch, Avila Pires, Guo, Gheshlaghi Azar (b15) 2020; Vol. 33
Srivastava, Hinton, Krizhevsky, Sutskever, Salakhutdinov (b42) 2014; 15
Xu, Niethammer (b57) 2019
Devries, Taylor (b10) 2017
Tomar, Bozorgtabar, Lortkipanidze, Vray, Rad, Thiran (b45) 2022
Henaff (b19) 2020
Ronneberger, Fischer, Brox (b39) 2015
Ochal, Patacchiola, Vazquez, Storkey, Wang (b35) 2023; 4
Liu, Li, Di Wu, Chen, Wu, Guo, Li (b29) 2022
Hoopes, Hoffmann, Greve, Fischl, Guttag, Dalca (b21) 2022
Ghiasi, Lin, Le (b14) 2018; 31
Tang, Gao, Lee, Han, Chen, Gao, Nath, Bermudez, Savona, Abramson, Bao, Lyu, Huo, Landman (b44) 2021; 69
Vincent, Larochelle, Bengio, Manzagol (b50) 2008
He, Ge, Qi, Chen, Wu, Coatrieux, Yang, Li (b17) 2022
Liu, Zhang, Hou, Mian, Wang, Zhang, Tang (b30) 2023; 35
Wang, Sun, Seidlitz, Bethlehem, Alexander-Bloch, Dorfschmidt, Li, Elison, Lin, Wang (b54) 2025
Zhang, Cisse, Dauphin, Lopez-Paz (b59) 2018
Zheng, Sui, Jiang, Che, Zhang, Yang, Li (b63) 2022; 44
Gella, Gangloff, Wendt, Tiede, Lang (b13) 2023
Wang, Wu, Chen, Sun, Lin, Li (b55) 2023; 18
Ding, Y., Yu, X., Yang, Y., 2021. Modeling the probabilistic distribution of unlabeled data for one-shot medical image segmentation. In: Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 35, pp. 1246–1254.
He, Li, Yang, Kong, Chen, Shu, Coatrieux, Dillenseger, Li (b18) 2020
Verma, Lamb, Beckham, Najafi, Mitliagkas, Lopez-Paz, Bengio (b49) 2019
Zhao (10.1016/j.media.2025.103637_b61) 2019; 24
Zhang (10.1016/j.media.2025.103637_b59) 2018
Pu (10.1016/j.media.2025.103637_b37) 2016
Vahdat (10.1016/j.media.2025.103637_b48) 2020; Vol. 33
Tu (10.1016/j.media.2025.103637_b47) 2022; 70
Liu (10.1016/j.media.2025.103637_b30) 2023; 35
Grill (10.1016/j.media.2025.103637_b15) 2020; Vol. 33
Liu (10.1016/j.media.2025.103637_b29) 2022
Zhong (10.1016/j.media.2025.103637_b64) 2017
He (10.1016/j.media.2025.103637_b18) 2020
Marcus (10.1016/j.media.2025.103637_b34) 2007; 19
Wang (10.1016/j.media.2025.103637_b51) 2020
Avants (10.1016/j.media.2025.103637_b3) 2008; 12
Kingma (10.1016/j.media.2025.103637_b26) 2014; Vol. 27
Sekuboyina (10.1016/j.media.2025.103637_b40) 2021; 73
Ahn (10.1016/j.media.2025.103637_b1) 2024; 132
Wang (10.1016/j.media.2025.103637_b53) 2019
Tomar (10.1016/j.media.2025.103637_b45) 2022
Zheng (10.1016/j.media.2025.103637_b63) 2022; 44
Tang (10.1016/j.media.2025.103637_b44) 2021; 69
Avants (10.1016/j.media.2025.103637_b4) 2009; 2
He (10.1016/j.media.2025.103637_b17) 2022
Kingma (10.1016/j.media.2025.103637_b27) 2014
Jiang (10.1016/j.media.2025.103637_b23) 2022; 41
10.1016/j.media.2025.103637_b7
Hoopes (10.1016/j.media.2025.103637_b21) 2022
Wang (10.1016/j.media.2025.103637_b55) 2023; 18
Zhao (10.1016/j.media.2025.103637_b62) 2017
Henaff (10.1016/j.media.2025.103637_b19) 2020
Wang (10.1016/j.media.2025.103637_b52) 2023
Chen (10.1016/j.media.2025.103637_b8) 2022; 82
Ghiasi (10.1016/j.media.2025.103637_b14) 2018; 31
Gella (10.1016/j.media.2025.103637_b13) 2023
Zhao (10.1016/j.media.2025.103637_b60) 2019
10.1016/j.media.2025.103637_b12
10.1016/j.media.2025.103637_b11
Kang (10.1016/j.media.2025.103637_b25) 2022; 78
Balakrishnan (10.1016/j.media.2025.103637_b5) 2019; 38
Lv (10.1016/j.media.2025.103637_b31) 2022; 41
Ma (10.1016/j.media.2025.103637_b32) 2024; 15
Verma (10.1016/j.media.2025.103637_b49) 2019
Hering (10.1016/j.media.2025.103637_b20) 2022; 42
Ochal (10.1016/j.media.2025.103637_b35) 2023; 4
Antonelli (10.1016/j.media.2025.103637_b2) 2022; 13
Han (10.1016/j.media.2025.103637_b16) 2022
Wang (10.1016/j.media.2025.103637_b56) 2020
Rezende (10.1016/j.media.2025.103637_b38) 2014
Liu (10.1016/j.media.2025.103637_b28) 2023
Madry (10.1016/j.media.2025.103637_b33) 2018
Srivastava (10.1016/j.media.2025.103637_b42) 2014; 15
Vincent (10.1016/j.media.2025.103637_b50) 2008
Wang (10.1016/j.media.2025.103637_b54) 2025
Cheng (10.1016/j.media.2025.103637_b9) 2024; 43
Ronneberger (10.1016/j.media.2025.103637_b39) 2015
Tomczak (10.1016/j.media.2025.103637_b46) 2018
Snell (10.1016/j.media.2025.103637_b41) 2017; 30
Yun (10.1016/j.media.2025.103637_b58) 2019
Billion Polak (10.1016/j.media.2025.103637_b6) 2024; 11
Joy (10.1016/j.media.2025.103637_b24) 2021
Oord (10.1016/j.media.2025.103637_b36) 2018
Isensee (10.1016/j.media.2025.103637_b22) 2021; 18
Xu (10.1016/j.media.2025.103637_b57) 2019
Devries (10.1016/j.media.2025.103637_b10) 2017
Talbot (10.1016/j.media.2025.103637_b43) 2020
References_xml – start-page: 8543
  year: 2019
  end-page: 8553
  ident: b60
  article-title: Data augmentation using learned transformations for one-shot medical image segmentation
  publication-title: CVPR
– volume: Vol. 33
  start-page: 21271
  year: 2020
  end-page: 21284
  ident: b15
  article-title: Bootstrap your own latent-a new approach to self-supervised learning
  publication-title: NeurIPS
– reference: Ding, Y., Yu, X., Yang, Y., 2021. Modeling the probabilistic distribution of unlabeled data for one-shot medical image segmentation. In: Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 35, pp. 1246–1254.
– volume: 15
  start-page: 1929
  year: 2014
  end-page: 1958
  ident: b42
  article-title: Dropout: A simple way to prevent neural networks from overfitting
  publication-title: J. Mach. Learn. Res.
– volume: Vol. 27
  year: 2014
  ident: b26
  article-title: Semi-supervised learning with deep generative models
  publication-title: NeurIPS
– year: 2023
  ident: b13
  article-title: Self-supervised variational autoencoder for unsupervised object counting from very high-resolution satellite imagery: Applications in dwelling extraction in FDP settlement areas
  publication-title: IEEE Trans. Geosci. Remote Sens.
– volume: 35
  start-page: 857
  year: 2023
  end-page: 876
  ident: b30
  article-title: Self-supervised learning: Generative or contrastive
  publication-title: IEEE Trans. Knowl. Data Eng.
– volume: 73
  year: 2021
  ident: b40
  article-title: Verse: a vertebrae labelling and segmentation benchmark for multi-detector CT images
  publication-title: Med. Image Anal.
– start-page: 234
  year: 2015
  end-page: 241
  ident: b39
  article-title: U-net: Convolutional networks for biomedical image segmentation
  publication-title: MICCAI
– volume: 12
  start-page: 26
  year: 2008
  end-page: 41
  ident: b3
  article-title: Symmetric diffeomorphic image registration with cross-correlation: evaluating automated labeling of elderly and neurodegenerative brain
  publication-title: Med. Image Anal.
– volume: 18
  start-page: 203
  year: 2021
  end-page: 211
  ident: b22
  article-title: Nnu-net: a self-configuring method for deep learning-based biomedical image segmentation
  publication-title: Nature Methods
– volume: 30
  year: 2017
  ident: b41
  article-title: Prototypical networks for few-shot learning
  publication-title: NeurIPS
– start-page: 4091
  year: 2017
  end-page: 4099
  ident: b62
  article-title: Learning hierarchical features from deep generative models
  publication-title: ICML
– year: 2018
  ident: b36
  article-title: Representation learning with contrastive predictive coding
– volume: 38
  start-page: 1788
  year: 2019
  end-page: 1800
  ident: b5
  article-title: VoxelMorph: a learning framework for deformable medical image registration
  publication-title: IEEE Trans. Med. Imaging
– year: 2017
  ident: b10
  article-title: Improved regularization of convolutional neural networks with cutout
– volume: 78
  year: 2022
  ident: b25
  article-title: Dual-stream pyramid registration network
  publication-title: Med. Image Anal.
– year: 2017
  ident: b64
  article-title: Random erasing data augmentation
– volume: 70
  start-page: 5954
  year: 2022
  end-page: 5966
  ident: b47
  article-title: Supervising the decoder of variational autoencoders to improve scientific utility
  publication-title: IEEE Trans. Signal Process.
– volume: 31
  year: 2018
  ident: b14
  article-title: Dropblock: A regularization method for convolutional networks
  publication-title: NeurIPS
– volume: 2
  start-page: 1
  year: 2009
  end-page: 35
  ident: b4
  article-title: Advanced normalization tools (ANTS)
  publication-title: Insight J
– start-page: 1278
  year: 2014
  end-page: 1286
  ident: b38
  article-title: Stochastic backpropagation and approximate inference in deep generative models
  publication-title: ICML
– volume: 82
  year: 2022
  ident: b8
  article-title: Transmorph: Transformer for unsupervised medical image registration
  publication-title: Med. Image Anal.
– volume: 42
  start-page: 697
  year: 2022
  end-page: 712
  ident: b20
  article-title: Learn2Reg: comprehensive multi-task medical image registration challenge, dataset and evaluation in the era of deep learning
  publication-title: IEEE Trans. Med. Imaging
– year: 2018
  ident: b59
  article-title: Mixup: Beyond empirical risk minimization
  publication-title: ICLR
– year: 2022
  ident: b21
  article-title: Learning the effect of registration hyperparameters with hypermorph
– start-page: 1214
  year: 2018
  end-page: 1223
  ident: b46
  article-title: VAE with a VampPrior
  publication-title: International Conference on Artificial Intelligence and Statistics
– start-page: 12275
  year: 2020
  end-page: 12284
  ident: b56
  article-title: Self-supervised equivariant attention mechanism for weakly supervised semantic segmentation
  publication-title: CVPR
– year: 2018
  ident: b33
  article-title: Towards deep learning models resistant to adversarial attacks
  publication-title: ICLR
– year: 2023
  ident: b28
  article-title: Contrastive registration for unsupervised medical image segmentation
  publication-title: IEEE Trans. Neural Netw. Learn. Syst.
– start-page: 420
  year: 2019
  end-page: 429
  ident: b57
  article-title: DeepAtlas: Joint semi-supervised learning of image registration and segmentation
  publication-title: Medical Image Computing and Computer Assisted Intervention–MICCAI 2019: 22nd International Conference, Shenzhen, China, October 13–17, 2019, Proceedings, Part II 22
– volume: 43
  year: 2024
  ident: b9
  article-title: Few-shot medical image segmentation via generating multiple representative descriptors
  publication-title: IEEE Trans. Med. Imaging
– volume: 132
  year: 2024
  ident: b1
  article-title: Data augmentation for low-level vision: CutBlur and mixture-of-augmentation
  publication-title: Int. J. Comput. Vis.
– start-page: 1998
  year: 2022
  end-page: 2008
  ident: b45
  article-title: Self-supervised generative style transfer for one-shot medical image segmentation
  publication-title: CVPR
– start-page: 8230
  year: 2022
  end-page: 8248
  ident: b16
  article-title: G-mixup: Graph data augmentation for graph classification
  publication-title: ICML
– year: 2014
  ident: b27
  article-title: Auto-encoding variational Bayes
  publication-title: ICLR
– start-page: 441
  year: 2022
  end-page: 458
  ident: b29
  article-title: AutoMix: Unveiling the power of mixup for stronger classifiers
  publication-title: ECCV
– start-page: 2360
  year: 2016
  end-page: 2368
  ident: b37
  article-title: Variational autoencoder for deep learning of images, labels and captions
  publication-title: NeurIPS
– start-page: 9162
  year: 2020
  end-page: 9171
  ident: b51
  article-title: LT-net: Label transfer by learning reversible voxel-wise correspondence for one-shot medical image segmentation
  publication-title: CVPR
– volume: 41
  start-page: 2021
  year: 2022
  end-page: 2032
  ident: b23
  article-title: One shot PACS: Patient specific anatomic context and shape prior aware recurrent registration-segmentation of longitudinal thoracic cone beam CTs
  publication-title: IEEE Trans. Med. Imaging
– volume: 44
  start-page: 5631
  year: 2022
  end-page: 5646
  ident: b63
  article-title: SymReg-GAN: Symmetric image registration with generative adversarial networks
  publication-title: IEEE Trans. Pattern Anal. Mach. Intell.
– year: 2025
  ident: b54
  article-title: A lifespan-generalizable skull-stripping model for magnetic resonance images that leverages prior knowledge from brain atlases
  publication-title: Nat. Biomed. Eng.
– volume: 18
  start-page: 1488
  year: 2023
  end-page: 1509
  ident: b55
  article-title: iBEAT V2.0: a multisite-applicable, deep learning-based pipeline for infant cerebral cortical surface reconstruction
  publication-title: Nat. Protoc.
– year: 2022
  ident: b17
  article-title: Learning better registration to learn better few-shot medical image segmentation: Authenticity, diversity, and robustness
  publication-title: IEEE Trans. Neural Netw. Learn. Syst.
– start-page: 4182
  year: 2020
  end-page: 4192
  ident: b19
  article-title: Data-efficient image recognition with contrastive predictive coding
  publication-title: ICML
– reference: Fan, X., Wang, X., Gao, J., Wang, J., Luo, Z., Liu, R., 2024. Bi-level Learning of Task-Specific Decoders for Joint Registration and One-Shot Medical Image Segmentation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 11726–11735.
– volume: Vol. 33
  start-page: 19667
  year: 2020
  end-page: 19679
  ident: b48
  article-title: NVAE: A deep hierarchical variational autoencoder
  publication-title: NeurIPS
– volume: 41
  start-page: 2788
  year: 2022
  end-page: 2802
  ident: b31
  article-title: Joint progressive and coarse-to-fine registration of brain MRI via deformation field integration and non-rigid feature fusion
  publication-title: IEEE Trans. Med. Imaging
– volume: 69
  year: 2021
  ident: b44
  article-title: High-resolution 3D abdominal segmentation with random patch network fusion
  publication-title: Med. Image Anal.
– start-page: 1096
  year: 2008
  end-page: 1103
  ident: b50
  article-title: Extracting and composing robust features with denoising autoencoders
  publication-title: ICML
– start-page: 6438
  year: 2019
  end-page: 6447
  ident: b49
  article-title: Manifold mixup: Better representations by interpolating hidden states
  publication-title: ICML
– reference: Butoi, V.I., Ortiz, J.J.G., Ma, T., Sabuncu, M.R., Guttag, J., Dalca, A.V., 2023. Universeg: Universal medical image segmentation. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 21438–21451.
– volume: 24
  start-page: 1394
  year: 2019
  end-page: 1404
  ident: b61
  article-title: Unsupervised 3D end-to-end medical image registration with volume tweening network
  publication-title: IEEE J. Biomed. Heal. Inform.
– start-page: 9197
  year: 2019
  end-page: 9206
  ident: b53
  article-title: Panet: Few-shot image semantic segmentation with prototype alignment
  publication-title: ICCV
– start-page: 6023
  year: 2019
  end-page: 6032
  ident: b58
  article-title: Cutmix: Regularization strategy to train strong classifiers with localizable features
  publication-title: ICCV
– volume: 15
  start-page: 654
  year: 2024
  ident: b32
  article-title: Segment anything in medical images
  publication-title: Nat. Commun.
– year: 2020
  ident: b43
  article-title: Supervised autoencoders learn robust joint factor models of neural activity
– volume: 13
  start-page: 4128
  year: 2022
  ident: b2
  article-title: The medical segmentation decathlon
  publication-title: Nat. Commun.
– volume: 4
  start-page: 1348
  year: 2023
  end-page: 1358
  ident: b35
  article-title: Few-shot learning with class imbalance
  publication-title: IEEE Trans. Artif. Intell.
– volume: 19
  start-page: 1498
  year: 2007
  end-page: 1507
  ident: b34
  article-title: Open access series of imaging studies (OASIS): cross-sectional MRI data in young, middle aged, nondemented, and demented older adults
  publication-title: J. Cogn. Neurosci.
– year: 2023
  ident: b52
  article-title: SAM-Med3D: Towards general-purpose segmentation models for volumetric medical images
– volume: 11
  start-page: 1
  year: 2024
  ident: b6
  article-title: Low-shot learning and class imbalance: a survey
  publication-title: J. Big Data
– year: 2021
  ident: b24
  article-title: Capturing label characteristics in VAEs
  publication-title: ICLR
– start-page: 770
  year: 2020
  end-page: 786
  ident: b18
  article-title: Deep complementary joint model for complex scene registration and few-shot segmentation on medical images
  publication-title: Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XVIII 16
– year: 2021
  ident: 10.1016/j.media.2025.103637_b24
  article-title: Capturing label characteristics in VAEs
– volume: 38
  start-page: 1788
  issue: 8
  year: 2019
  ident: 10.1016/j.media.2025.103637_b5
  article-title: VoxelMorph: a learning framework for deformable medical image registration
  publication-title: IEEE Trans. Med. Imaging
  doi: 10.1109/TMI.2019.2897538
– year: 2018
  ident: 10.1016/j.media.2025.103637_b33
  article-title: Towards deep learning models resistant to adversarial attacks
– year: 2017
  ident: 10.1016/j.media.2025.103637_b10
– year: 2018
  ident: 10.1016/j.media.2025.103637_b59
  article-title: Mixup: Beyond empirical risk minimization
– year: 2022
  ident: 10.1016/j.media.2025.103637_b21
– volume: 42
  start-page: 697
  issue: 3
  year: 2022
  ident: 10.1016/j.media.2025.103637_b20
  article-title: Learn2Reg: comprehensive multi-task medical image registration challenge, dataset and evaluation in the era of deep learning
  publication-title: IEEE Trans. Med. Imaging
  doi: 10.1109/TMI.2022.3213983
– start-page: 4091
  year: 2017
  ident: 10.1016/j.media.2025.103637_b62
  article-title: Learning hierarchical features from deep generative models
– year: 2018
  ident: 10.1016/j.media.2025.103637_b36
– start-page: 1998
  year: 2022
  ident: 10.1016/j.media.2025.103637_b45
  article-title: Self-supervised generative style transfer for one-shot medical image segmentation
– start-page: 9162
  year: 2020
  ident: 10.1016/j.media.2025.103637_b51
  article-title: LT-net: Label transfer by learning reversible voxel-wise correspondence for one-shot medical image segmentation
– volume: 132
  issue: 6
  year: 2024
  ident: 10.1016/j.media.2025.103637_b1
  article-title: Data augmentation for low-level vision: CutBlur and mixture-of-augmentation
  publication-title: Int. J. Comput. Vis.
  doi: 10.1007/s11263-023-01970-z
– year: 2022
  ident: 10.1016/j.media.2025.103637_b17
  article-title: Learning better registration to learn better few-shot medical image segmentation: Authenticity, diversity, and robustness
  publication-title: IEEE Trans. Neural Netw. Learn. Syst.
– start-page: 4182
  year: 2020
  ident: 10.1016/j.media.2025.103637_b19
  article-title: Data-efficient image recognition with contrastive predictive coding
– volume: 4
  start-page: 1348
  issue: 5
  year: 2023
  ident: 10.1016/j.media.2025.103637_b35
  article-title: Few-shot learning with class imbalance
  publication-title: IEEE Trans. Artif. Intell.
  doi: 10.1109/TAI.2023.3298303
– volume: 82
  year: 2022
  ident: 10.1016/j.media.2025.103637_b8
  article-title: Transmorph: Transformer for unsupervised medical image registration
  publication-title: Med. Image Anal.
  doi: 10.1016/j.media.2022.102615
– volume: 24
  start-page: 1394
  issue: 5
  year: 2019
  ident: 10.1016/j.media.2025.103637_b61
  article-title: Unsupervised 3D end-to-end medical image registration with volume tweening network
  publication-title: IEEE J. Biomed. Heal. Inform.
  doi: 10.1109/JBHI.2019.2951024
– volume: 18
  start-page: 1488
  issue: 5
  year: 2023
  ident: 10.1016/j.media.2025.103637_b55
  article-title: iBEAT V2.0: a multisite-applicable, deep learning-based pipeline for infant cerebral cortical surface reconstruction
  publication-title: Nat. Protoc.
  doi: 10.1038/s41596-023-00806-x
– volume: 2
  start-page: 1
  issue: 365
  year: 2009
  ident: 10.1016/j.media.2025.103637_b4
  article-title: Advanced normalization tools (ANTS)
  publication-title: Insight J
– start-page: 770
  year: 2020
  ident: 10.1016/j.media.2025.103637_b18
  article-title: Deep complementary joint model for complex scene registration and few-shot segmentation on medical images
– start-page: 2360
  year: 2016
  ident: 10.1016/j.media.2025.103637_b37
  article-title: Variational autoencoder for deep learning of images, labels and captions
– volume: 35
  start-page: 857
  issue: 1
  year: 2023
  ident: 10.1016/j.media.2025.103637_b30
  article-title: Self-supervised learning: Generative or contrastive
  publication-title: IEEE Trans. Knowl. Data Eng.
– start-page: 8230
  year: 2022
  ident: 10.1016/j.media.2025.103637_b16
  article-title: G-mixup: Graph data augmentation for graph classification
– volume: 73
  year: 2021
  ident: 10.1016/j.media.2025.103637_b40
  article-title: Verse: a vertebrae labelling and segmentation benchmark for multi-detector CT images
  publication-title: Med. Image Anal.
  doi: 10.1016/j.media.2021.102166
– volume: 70
  start-page: 5954
  year: 2022
  ident: 10.1016/j.media.2025.103637_b47
  article-title: Supervising the decoder of variational autoencoders to improve scientific utility
  publication-title: IEEE Trans. Signal Process.
  doi: 10.1109/TSP.2022.3230329
– year: 2023
  ident: 10.1016/j.media.2025.103637_b28
  article-title: Contrastive registration for unsupervised medical image segmentation
  publication-title: IEEE Trans. Neural Netw. Learn. Syst.
– year: 2025
  ident: 10.1016/j.media.2025.103637_b54
  article-title: A lifespan-generalizable skull-stripping model for magnetic resonance images that leverages prior knowledge from brain atlases
  publication-title: Nat. Biomed. Eng.
  doi: 10.1038/s41551-024-01337-w
– volume: Vol. 27
  year: 2014
  ident: 10.1016/j.media.2025.103637_b26
  article-title: Semi-supervised learning with deep generative models
– ident: 10.1016/j.media.2025.103637_b7
  doi: 10.1109/ICCV51070.2023.01960
– start-page: 1214
  year: 2018
  ident: 10.1016/j.media.2025.103637_b46
  article-title: VAE with a VampPrior
– ident: 10.1016/j.media.2025.103637_b12
  doi: 10.1109/CVPR52733.2024.01114
– volume: 31
  year: 2018
  ident: 10.1016/j.media.2025.103637_b14
  article-title: Dropblock: A regularization method for convolutional networks
  publication-title: NeurIPS
– start-page: 8543
  year: 2019
  ident: 10.1016/j.media.2025.103637_b60
  article-title: Data augmentation using learned transformations for one-shot medical image segmentation
– volume: 41
  start-page: 2021
  issue: 8
  year: 2022
  ident: 10.1016/j.media.2025.103637_b23
  article-title: One shot PACS: Patient specific anatomic context and shape prior aware recurrent registration-segmentation of longitudinal thoracic cone beam CTs
  publication-title: IEEE Trans. Med. Imaging
  doi: 10.1109/TMI.2022.3154934
– volume: 30
  year: 2017
  ident: 10.1016/j.media.2025.103637_b41
  article-title: Prototypical networks for few-shot learning
– volume: 19
  start-page: 1498
  issue: 9
  year: 2007
  ident: 10.1016/j.media.2025.103637_b34
  article-title: Open access series of imaging studies (OASIS): cross-sectional MRI data in young, middle aged, nondemented, and demented older adults
  publication-title: J. Cogn. Neurosci.
  doi: 10.1162/jocn.2007.19.9.1498
– volume: 15
  start-page: 1929
  issue: 56
  year: 2014
  ident: 10.1016/j.media.2025.103637_b42
  article-title: Dropout: A simple way to prevent neural networks from overfitting
  publication-title: J. Mach. Learn. Res.
– volume: 13
  start-page: 4128
  issue: 1
  year: 2022
  ident: 10.1016/j.media.2025.103637_b2
  article-title: The medical segmentation decathlon
  publication-title: Nat. Commun.
  doi: 10.1038/s41467-022-30695-9
– start-page: 1096
  year: 2008
  ident: 10.1016/j.media.2025.103637_b50
  article-title: Extracting and composing robust features with denoising autoencoders
– year: 2023
  ident: 10.1016/j.media.2025.103637_b52
– start-page: 420
  year: 2019
  ident: 10.1016/j.media.2025.103637_b57
  article-title: DeepAtlas: Joint semi-supervised learning of image registration and segmentation
– ident: 10.1016/j.media.2025.103637_b11
  doi: 10.1609/aaai.v35i2.16212
– volume: 18
  start-page: 203
  issue: 2
  year: 2021
  ident: 10.1016/j.media.2025.103637_b22
  article-title: Nnu-net: a self-configuring method for deep learning-based biomedical image segmentation
  publication-title: Nature Methods
  doi: 10.1038/s41592-020-01008-z
– year: 2017
  ident: 10.1016/j.media.2025.103637_b64
– year: 2020
  ident: 10.1016/j.media.2025.103637_b43
– volume: 43
  issue: 6
  year: 2024
  ident: 10.1016/j.media.2025.103637_b9
  article-title: Few-shot medical image segmentation via generating multiple representative descriptors
  publication-title: IEEE Trans. Med. Imaging
  doi: 10.1109/TMI.2024.3358295
– start-page: 6438
  year: 2019
  ident: 10.1016/j.media.2025.103637_b49
  article-title: Manifold mixup: Better representations by interpolating hidden states
– volume: 11
  start-page: 1
  issue: 1
  year: 2024
  ident: 10.1016/j.media.2025.103637_b6
  article-title: Low-shot learning and class imbalance: a survey
  publication-title: J. Big Data
  doi: 10.1186/s40537-023-00851-z
– volume: 41
  start-page: 2788
  issue: 10
  year: 2022
  ident: 10.1016/j.media.2025.103637_b31
  article-title: Joint progressive and coarse-to-fine registration of brain MRI via deformation field integration and non-rigid feature fusion
  publication-title: IEEE Trans. Med. Imaging
  doi: 10.1109/TMI.2022.3170879
– start-page: 9197
  year: 2019
  ident: 10.1016/j.media.2025.103637_b53
  article-title: Panet: Few-shot image semantic segmentation with prototype alignment
– start-page: 234
  year: 2015
  ident: 10.1016/j.media.2025.103637_b39
  article-title: U-net: Convolutional networks for biomedical image segmentation
– volume: Vol. 33
  start-page: 21271
  year: 2020
  ident: 10.1016/j.media.2025.103637_b15
  article-title: Bootstrap your own latent-a new approach to self-supervised learning
– start-page: 6023
  year: 2019
  ident: 10.1016/j.media.2025.103637_b58
  article-title: Cutmix: Regularization strategy to train strong classifiers with localizable features
– volume: Vol. 33
  start-page: 19667
  year: 2020
  ident: 10.1016/j.media.2025.103637_b48
  article-title: NVAE: A deep hierarchical variational autoencoder
– volume: 69
  year: 2021
  ident: 10.1016/j.media.2025.103637_b44
  article-title: High-resolution 3D abdominal segmentation with random patch network fusion
  publication-title: Med. Image Anal.
  doi: 10.1016/j.media.2020.101894
– start-page: 1278
  year: 2014
  ident: 10.1016/j.media.2025.103637_b38
  article-title: Stochastic backpropagation and approximate inference in deep generative models
– volume: 44
  start-page: 5631
  issue: 09
  year: 2022
  ident: 10.1016/j.media.2025.103637_b63
  article-title: SymReg-GAN: Symmetric image registration with generative adversarial networks
  publication-title: IEEE Trans. Pattern Anal. Mach. Intell.
– start-page: 12275
  year: 2020
  ident: 10.1016/j.media.2025.103637_b56
  article-title: Self-supervised equivariant attention mechanism for weakly supervised semantic segmentation
– volume: 12
  start-page: 26
  issue: 1
  year: 2008
  ident: 10.1016/j.media.2025.103637_b3
  article-title: Symmetric diffeomorphic image registration with cross-correlation: evaluating automated labeling of elderly and neurodegenerative brain
  publication-title: Med. Image Anal.
  doi: 10.1016/j.media.2007.06.004
– year: 2023
  ident: 10.1016/j.media.2025.103637_b13
  article-title: Self-supervised variational autoencoder for unsupervised object counting from very high-resolution satellite imagery: Applications in dwelling extraction in FDP settlement areas
  publication-title: IEEE Trans. Geosci. Remote Sens.
– volume: 78
  year: 2022
  ident: 10.1016/j.media.2025.103637_b25
  article-title: Dual-stream pyramid registration network
  publication-title: Med. Image Anal.
  doi: 10.1016/j.media.2022.102379
– year: 2014
  ident: 10.1016/j.media.2025.103637_b27
  article-title: Auto-encoding variational Bayes
– start-page: 441
  year: 2022
  ident: 10.1016/j.media.2025.103637_b29
  article-title: AutoMix: Unveiling the power of mixup for stronger classifiers
– volume: 15
  start-page: 654
  issue: 1
  year: 2024
  ident: 10.1016/j.media.2025.103637_b32
  article-title: Segment anything in medical images
  publication-title: Nat. Commun.
  doi: 10.1038/s41467-024-44824-z
SSID ssj0007440
Score 2.4524922
Snippet Few-shot medical image segmentation typically uses a joint model for registration and segmentation. The registration model aligns a labeled atlas with...
SourceID proquest
pubmed
crossref
elsevier
SourceType Aggregation Database
Index Database
Publisher
StartPage 103637
SubjectTerms Algorithms
Autoencoder
Few-shot learning
Humans
Image Interpretation, Computer-Assisted - methods
Image Processing, Computer-Assisted - methods
Image reconstruction
Magnetic Resonance Imaging
Medical image segmentation
Supervised Machine Learning
Variational autoencoder
Title Efficient few-shot medical image segmentation via self-supervised variational autoencoder
URI https://dx.doi.org/10.1016/j.media.2025.103637
https://www.ncbi.nlm.nih.gov/pubmed/40449308
https://www.proquest.com/docview/3214326016
Volume 104
WOSCitedRecordID wos001502478000002&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: PRVESC
  databaseName: Elsevier SD Freedom Collection Journals 2021
  customDbUrl:
  eissn: 1361-8423
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0007440
  issn: 1361-8415
  databaseCode: AIEXJ
  dateStart: 20161201
  isFulltext: true
  titleUrlDefault: https://www.sciencedirect.com
  providerName: Elsevier
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1Lb9QwELZoixAcEJRHl0cVJMQFvIoT5-FjhbYCtCwctlI4Wd7EaXdFk22zWfrzGb-yaWlROXCJolGcRDNfJmPP-BuE3pJE-kwWDItCMExDkWAxEwX2cxGxCOZiqdQkruNkMkmzjH23GfxGtxNIqiq9uGDL_2pqkIGx1dbZfzB3d1MQwDkYHY5gdjjeyvAjTQqhOZfkL9yc1CuTQFfsGqeqQqeRx6d2x1H1fj0XIPhZ4qZdKrfRQAC6hvmzWyMU7apWXJeFreJ1zZ8u3VFYZpPNKnTdat8uqsVcXpWCto6dLJs7yaLdFAfNzejaXmcXJYKoK4lzfjSMCU6p2Uo8lNfInPP1ac99EpVWTq717GaRYTHUG2qG6pnDzdWXebQn3_jh0XjMp6Ns-m55hlWLMZWKt_1WttBOkEQMXODOwedR9qX7cSuuRLNNz7ynI6nS5YB_PPemQOamiYoOWKaP0EM70_AODEIeozuy2kUPevyTu-jeV1tZ8QT96GDjOdh4FjaeNrLXh40HsPGuwMbrwcbrweYpOjocTT9-wrbtBs5DQlaYwedJIVAscklkwgJJCcSpEYujoswDUaQ09WU5S0VJGM1DJuA3kc8gDhRlQuOYhc_QdlVXcg95QU6TWKpUP-g2SImI8oCEkR_mfsGIFAP0wemQLw27CndlhwuuVc6VyrlR-QDFTs_cBogm8OOAkr8PfOOswsF9qpyYqGTdNlz16Qo1rd4APTfm6t6E-pSy0E9f3GL0S3R_8x28Qtur81a-Rnfz9WrenO-jrSRL9y3efgPakZtT
linkProvider Elsevier
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=Efficient+few-shot+medical+image+segmentation+via+self-supervised+variational+autoencoder&rft.jtitle=Medical+image+analysis&rft.au=Zhou%2C+Yanjie&rft.au=Zhou%2C+Feng&rft.au=Xi%2C+Fengjun&rft.au=Liu%2C+Yong&rft.date=2025-08-01&rft.issn=1361-8423&rft.eissn=1361-8423&rft.volume=104&rft.spage=103637&rft_id=info:doi/10.1016%2Fj.media.2025.103637&rft.externalDBID=NO_FULL_TEXT
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1361-8415&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1361-8415&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1361-8415&client=summon