AttriMIL: Revisiting attention-based multiple instance learning for whole-slide pathological image classification from a perspective of instance attributes

Multiple instance learning (MIL) is a powerful approach for whole-slide pathological image (WSI) analysis, particularly suited for processing gigapixel-resolution images with slide-level labels. Recent attention-based MIL architectures have significantly advanced weakly supervised WSI classification...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Medical image analysis Ročník 103; s. 103631
Hlavní autori: Cai, Linghan, Huang, Shenjin, Zhang, Ye, Lu, Jinpeng, Zhang, Yongbing
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Netherlands Elsevier B.V 01.07.2025
Predmet:
ISSN:1361-8415, 1361-8423, 1361-8423
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract Multiple instance learning (MIL) is a powerful approach for whole-slide pathological image (WSI) analysis, particularly suited for processing gigapixel-resolution images with slide-level labels. Recent attention-based MIL architectures have significantly advanced weakly supervised WSI classification, facilitating both clinical diagnosis and localization of disease-positive regions. However, these methods often face challenges in differentiating between instances, leading to tissue misidentification and a potential degradation in classification performance. To address these limitations, we propose AttriMIL, an attribute-aware multiple instance learning framework. By dissecting the computational flow of attention-based MIL models, we introduce a multi-branch attribute scoring mechanism that quantifies the pathological attributes of individual instances. Leveraging these quantified attributes, we further establish region-wise and slide-wise attribute constraints to dynamically model instance correlations both within and across slides during training. These constraints encourage the network to capture intrinsic spatial patterns and semantic similarities between image patches, thereby enhancing its ability to distinguish subtle tissue variations and sensitivity to challenging instances. To fully exploit the two constraints, we further develop a pathology adaptive learning technique to optimize pre-trained feature extractors, enabling the model to efficiently gather task-specific features. Extensive experiments on five public datasets demonstrate that AttriMIL consistently outperforms state-of-the-art methods across various dimensions, including bag classification accuracy, generalization ability, and disease-positive region localization. The implementation code is available at https://github.com/MedCAI/AttriMIL. [Display omitted] •We present AttriMIL with multi-branch attribute scoring.•Region-wise attribute constraint uses spatial patterns to boost sensitivity.•Slide-wise attribute constraint models instance correlations across WSIs.•Pathology adaptive learning exploits the two constraints for refined features.•Experiments show AttriMIL’s superiority, achieving state-of-the-art performance.
AbstractList Multiple instance learning (MIL) is a powerful approach for whole-slide pathological image (WSI) analysis, particularly suited for processing gigapixel-resolution images with slide-level labels. Recent attention-based MIL architectures have significantly advanced weakly supervised WSI classification, facilitating both clinical diagnosis and localization of disease-positive regions. However, these methods often face challenges in differentiating between instances, leading to tissue misidentification and a potential degradation in classification performance. To address these limitations, we propose AttriMIL, an attribute-aware multiple instance learning framework. By dissecting the computational flow of attention-based MIL models, we introduce a multi-branch attribute scoring mechanism that quantifies the pathological attributes of individual instances. Leveraging these quantified attributes, we further establish region-wise and slide-wise attribute constraints to dynamically model instance correlations both within and across slides during training. These constraints encourage the network to capture intrinsic spatial patterns and semantic similarities between image patches, thereby enhancing its ability to distinguish subtle tissue variations and sensitivity to challenging instances. To fully exploit the two constraints, we further develop a pathology adaptive learning technique to optimize pre-trained feature extractors, enabling the model to efficiently gather task-specific features. Extensive experiments on five public datasets demonstrate that AttriMIL consistently outperforms state-of-the-art methods across various dimensions, including bag classification accuracy, generalization ability, and disease-positive region localization. The implementation code is available at https://github.com/MedCAI/AttriMIL. [Display omitted] •We present AttriMIL with multi-branch attribute scoring.•Region-wise attribute constraint uses spatial patterns to boost sensitivity.•Slide-wise attribute constraint models instance correlations across WSIs.•Pathology adaptive learning exploits the two constraints for refined features.•Experiments show AttriMIL’s superiority, achieving state-of-the-art performance.
Multiple instance learning (MIL) is a powerful approach for whole-slide pathological image (WSI) analysis, particularly suited for processing gigapixel-resolution images with slide-level labels. Recent attention-based MIL architectures have significantly advanced weakly supervised WSI classification, facilitating both clinical diagnosis and localization of disease-positive regions. However, these methods often face challenges in differentiating between instances, leading to tissue misidentification and a potential degradation in classification performance. To address these limitations, we propose AttriMIL, an attribute-aware multiple instance learning framework. By dissecting the computational flow of attention-based MIL models, we introduce a multi-branch attribute scoring mechanism that quantifies the pathological attributes of individual instances. Leveraging these quantified attributes, we further establish region-wise and slide-wise attribute constraints to dynamically model instance correlations both within and across slides during training. These constraints encourage the network to capture intrinsic spatial patterns and semantic similarities between image patches, thereby enhancing its ability to distinguish subtle tissue variations and sensitivity to challenging instances. To fully exploit the two constraints, we further develop a pathology adaptive learning technique to optimize pre-trained feature extractors, enabling the model to efficiently gather task-specific features. Extensive experiments on five public datasets demonstrate that AttriMIL consistently outperforms state-of-the-art methods across various dimensions, including bag classification accuracy, generalization ability, and disease-positive region localization. The implementation code is available at https://github.com/MedCAI/AttriMIL.
Multiple instance learning (MIL) is a powerful approach for whole-slide pathological image (WSI) analysis, particularly suited for processing gigapixel-resolution images with slide-level labels. Recent attention-based MIL architectures have significantly advanced weakly supervised WSI classification, facilitating both clinical diagnosis and localization of disease-positive regions. However, these methods often face challenges in differentiating between instances, leading to tissue misidentification and a potential degradation in classification performance. To address these limitations, we propose AttriMIL, an attribute-aware multiple instance learning framework. By dissecting the computational flow of attention-based MIL models, we introduce a multi-branch attribute scoring mechanism that quantifies the pathological attributes of individual instances. Leveraging these quantified attributes, we further establish region-wise and slide-wise attribute constraints to dynamically model instance correlations both within and across slides during training. These constraints encourage the network to capture intrinsic spatial patterns and semantic similarities between image patches, thereby enhancing its ability to distinguish subtle tissue variations and sensitivity to challenging instances. To fully exploit the two constraints, we further develop a pathology adaptive learning technique to optimize pre-trained feature extractors, enabling the model to efficiently gather task-specific features. Extensive experiments on five public datasets demonstrate that AttriMIL consistently outperforms state-of-the-art methods across various dimensions, including bag classification accuracy, generalization ability, and disease-positive region localization. The implementation code is available at https://github.com/MedCAI/AttriMIL.Multiple instance learning (MIL) is a powerful approach for whole-slide pathological image (WSI) analysis, particularly suited for processing gigapixel-resolution images with slide-level labels. Recent attention-based MIL architectures have significantly advanced weakly supervised WSI classification, facilitating both clinical diagnosis and localization of disease-positive regions. However, these methods often face challenges in differentiating between instances, leading to tissue misidentification and a potential degradation in classification performance. To address these limitations, we propose AttriMIL, an attribute-aware multiple instance learning framework. By dissecting the computational flow of attention-based MIL models, we introduce a multi-branch attribute scoring mechanism that quantifies the pathological attributes of individual instances. Leveraging these quantified attributes, we further establish region-wise and slide-wise attribute constraints to dynamically model instance correlations both within and across slides during training. These constraints encourage the network to capture intrinsic spatial patterns and semantic similarities between image patches, thereby enhancing its ability to distinguish subtle tissue variations and sensitivity to challenging instances. To fully exploit the two constraints, we further develop a pathology adaptive learning technique to optimize pre-trained feature extractors, enabling the model to efficiently gather task-specific features. Extensive experiments on five public datasets demonstrate that AttriMIL consistently outperforms state-of-the-art methods across various dimensions, including bag classification accuracy, generalization ability, and disease-positive region localization. The implementation code is available at https://github.com/MedCAI/AttriMIL.
ArticleNumber 103631
Author Lu, Jinpeng
Cai, Linghan
Huang, Shenjin
Zhang, Yongbing
Zhang, Ye
Author_xml – sequence: 1
  givenname: Linghan
  orcidid: 0000-0002-7931-7697
  surname: Cai
  fullname: Cai, Linghan
  email: cailh@stu.hit.edu.cn
  organization: School of Computer Science and Technology, Harbin Institute of Technology (Shenzhen), Shenzhen, 518055, China
– sequence: 2
  givenname: Shenjin
  surname: Huang
  fullname: Huang, Shenjin
  organization: Faculty of Computing, Harbin Institute of Technology, Harbin, 150001, China
– sequence: 3
  givenname: Ye
  orcidid: 0009-0003-7254-5026
  surname: Zhang
  fullname: Zhang, Ye
  organization: School of Computer Science and Technology, Harbin Institute of Technology (Shenzhen), Shenzhen, 518055, China
– sequence: 4
  givenname: Jinpeng
  surname: Lu
  fullname: Lu, Jinpeng
  organization: School of Science, Harbin Institute of Technology (Shenzhen), Shenzhen, 518055, China
– sequence: 5
  givenname: Yongbing
  orcidid: 0000-0003-3320-2904
  surname: Zhang
  fullname: Zhang, Yongbing
  email: ybzhang08@hit.edu.cn
  organization: School of Computer Science and Technology, Harbin Institute of Technology (Shenzhen), Shenzhen, 518055, China
BackLink https://www.ncbi.nlm.nih.gov/pubmed/40381256$$D View this record in MEDLINE/PubMed
BookMark eNp9kUtr3DAUhUVJaB7tLygULbvxVA-_ptBFCE0amBAI7VrI8tX0DrLlSvKE_pb-2cqZpIUuspJ0-c5B95wzcjT6EQh5x9mKM15_3K0G6FGvBBNVnsha8lfklMuaF20p5NHfO69OyFmMO8ZYU5bsNTkpmWy5qOpT8vsipYC3N5tP9B72GDHhuKU6JRgT-rHodISeDrNLODmgOMakRwPUgQ7jglof6MMP76CIDnugk0755bdotKM46C1Q43SMaPNksaQ2-IFqOkGIE5iEe6De_nPWy4e6OUF8Q46tdhHePp3n5PvVl2-XX4vN3fXN5cWmMCWrU7FuDHDZNl3DGmtAdK2U2og1b0XJKtH1AK1ljYTWlDXveGOF5LUpecek1kLLc_Lh4DsF_3OGmNSA0YBzegQ_RyUFq-palA3L6PsndO5y-moKecXwSz0HmoH1ATDBxxjAKoPpce8UNDrFmVrKUzv1WJ5aylOH8rJW_qd9tn9Z9fmgghzRHiGoaBBykj2GHK_qPb6o_wP9ibcP
CitedBy_id crossref_primary_10_1002_jemt_70069
Cites_doi 10.1109/TCSVT.2025.3528625
10.1016/j.media.2023.102748
10.1109/CVPR52733.2024.01078
10.1109/CVPR.2016.266
10.1016/j.media.2024.103197
10.1038/s41551-020-00682-w
10.1016/j.media.2022.102559
10.1109/CVPR52733.2024.01076
10.1016/j.media.2023.102763
10.1164/ajrccm.151.6.7767535
10.1109/CVPR52688.2022.01824
10.1016/j.patcog.2017.10.009
10.1109/TMI.2018.2867350
10.1109/CVPR52729.2023.00720
10.1109/CVPR.2018.00678
10.1016/j.cell.2023.01.035
10.1609/aaai.v37i1.25160
10.1109/ICCVW60793.2023.00263
10.1109/TCSVT.2024.3400876
10.1109/CVPRW63382.2024.00683
10.1016/j.cmpb.2024.108161
10.3389/fonc.2021.759007
10.2307/2531595
10.1609/aaai.v36i2.20051
10.1146/annurev-pathol-011811-120902
10.1001/jama.2017.14585
10.1109/TMI.2023.3264781
10.1109/TMI.2023.3313252
10.1016/j.jpi.2023.100357
10.1109/CVPR52733.2024.01067
10.1109/ICCV51070.2023.00377
10.1109/CVPR52688.2022.01567
10.1109/CVPR.2016.90
10.1016/j.media.2024.103109
10.1016/j.media.2025.103456
10.1016/j.media.2024.103294
10.1109/CVPR46437.2021.01409
10.1109/TMI.2022.3157983
10.1109/TMI.2022.3176598
10.1016/j.compmedimag.2024.102337
10.1016/S0004-3702(97)00043-X
10.1038/s41591-019-0508-1
10.1038/s41591-024-02857-3
10.1016/j.media.2020.101813
ContentType Journal Article
Copyright 2025
Copyright © 2025. Published by Elsevier B.V.
Copyright_xml – notice: 2025
– notice: Copyright © 2025. Published by Elsevier B.V.
DBID 6I.
AAFTH
AAYXX
CITATION
CGR
CUY
CVF
ECM
EIF
NPM
7X8
DOI 10.1016/j.media.2025.103631
DatabaseName ScienceDirect Open Access Titles
Elsevier:ScienceDirect:Open Access
CrossRef
Medline
MEDLINE
MEDLINE (Ovid)
MEDLINE
MEDLINE
PubMed
MEDLINE - Academic
DatabaseTitle CrossRef
MEDLINE
Medline Complete
MEDLINE with Full Text
PubMed
MEDLINE (Ovid)
MEDLINE - Academic
DatabaseTitleList
MEDLINE
MEDLINE - Academic
Database_xml – sequence: 1
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 2
  dbid: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Medicine
Engineering
EISSN 1361-8423
ExternalDocumentID 40381256
10_1016_j_media_2025_103631
S1361841525001781
Genre Journal Article
GroupedDBID ---
--K
--M
.~1
0R~
1B1
1~.
1~5
29M
4.4
457
4G.
53G
5GY
5VS
6I.
7-5
71M
8P~
AAEDT
AAEDW
AAFTH
AAIKJ
AAKOC
AALRI
AAOAW
AAQFI
AAQXK
AATTM
AAXKI
AAXUO
AAYFN
AAYWO
ABBOA
ABBQC
ABJNI
ABMAC
ABMZM
ABWVN
ABXDB
ACDAQ
ACGFS
ACIEU
ACIUM
ACIWK
ACLOT
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
AGHFR
AGQPQ
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
EFKBS
EFLBG
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-
ROL
RPZ
SDF
SDG
SDP
SEL
SES
SEW
SPC
SPCBC
SSH
SST
SSV
SSZ
T5K
TEORI
UHS
~G-
~HD
9DU
AAYXX
CITATION
AGCQF
AGRNS
CGR
CUY
CVF
ECM
EIF
NPM
RIG
7X8
ID FETCH-LOGICAL-c406t-97ce1387b707fce2b833ac291824052bdee8f073e8c461b17f2316c41b03aa2a3
ISICitedReferencesCount 1
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001495110600001&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 23:08:11 EDT 2025
Sat May 31 02:14:10 EDT 2025
Sat Nov 29 07:06:06 EST 2025
Tue Nov 18 21:21:38 EST 2025
Sat Nov 15 16:52:37 EST 2025
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Keywords Pathology attribute constraint
Pathological image analysis
Attribute scoring mechanism
Multiple instance learning
Pathology adaptive learning
Language English
License This is an open access article under the CC BY license.
Copyright © 2025. Published by Elsevier B.V.
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c406t-97ce1387b707fce2b833ac291824052bdee8f073e8c461b17f2316c41b03aa2a3
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ORCID 0000-0003-3320-2904
0000-0002-7931-7697
0009-0003-7254-5026
OpenAccessLink https://dx.doi.org/10.1016/j.media.2025.103631
PMID 40381256
PQID 3205662470
PQPubID 23479
ParticipantIDs proquest_miscellaneous_3205662470
pubmed_primary_40381256
crossref_citationtrail_10_1016_j_media_2025_103631
crossref_primary_10_1016_j_media_2025_103631
elsevier_sciencedirect_doi_10_1016_j_media_2025_103631
PublicationCentury 2000
PublicationDate 2025-07-01
PublicationDateYYYYMMDD 2025-07-01
PublicationDate_xml – month: 07
  year: 2025
  text: 2025-07-01
  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 Sun, Q., Jiang, D., Li, J., Yan, R., He, Y., Guan, T., Cheng, Z., 2024. NcIEMIL: Rethinking Decoupled Multiple Instance Learning Framework for Histopathological Slide Classification. In: Medical Imaging with Deep Learning.
Marquette, Copin, Wallet, Neviere, Saulnier, Mathieu, Durocher, Ramon, Tonnel (b44) 1995; 151
Ramon, J., 2000. Multi instance neural networks. In: ML-2000 Workshop Attribute-Value and Relational Learning.
Bandi, Geessink, Manson, Van Dijk, Balkenhol, Hermsen, Bejnordi, Lee, Paeng, Zhong (b2) 2018; 38
Barbano, Perlo, Tartaglione, Fiandrotti, Bertero, Cassoni, Grangetto (b3) 2021
Hou, L., Samaras, D., Kurc, T.M., Gao, Y., Davis, J.E., Saltz, J.H., 2016. Patch-based convolutional neural network for whole slide tissue image classification. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. pp. 2424–2433.
Chen, Lu, Shaban, Chen, Chen, Williamson, Mahmood (b10) 2021
Grill, Strub, Altché, Tallec, Richemond, Buchatskaya, Doersch, Avila Pires, Guo, Gheshlaghi Azar (b22) 2020; 33
Yu, Wu, Ming, Deng, Li, Ou, He, Wang, Zhang, Wang (b73) 2023; 85
Chen, R.J., Chen, C., Li, Y., Chen, T.Y., Trister, A.D., Krishnan, R.G., Mahmood, F., 2022. Scaling vision transformers to gigapixel images via hierarchical self-supervised learning. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 16144–16155.
Ilse, Tomczak, Welling (b29) 2018
Ioffe (b30) 2015
He, X., Li, C., Zhang, P., Yang, J., Wang, X.E., 2023. Parameter-efficient model adaptation for vision transformers. In: Proceedings of the AAAI Conference on Artificial Intelligence. 37, (1), pp. 817–825.
Xu, Wang, Shi, Qin, Zhang, Liu, Madabhushi, Gao, Cong, Lu (b68) 2025
Ulyanov, Vedaldi, Lempitsky (b61) 2016
Jin, Luo, Lin, Hou, Chen (b31) 2024
Sultani, W., Chen, C., Shah, M., 2018. Real-world anomaly detection in surveillance videos. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. pp. 6479–6488.
Paszke, Gross, Massa, Lerer, Bradbury, Chanan, Killeen, Lin, Gimelshein, Antiga (b46) 2019; 32
Tan, Le (b56) 2021
Swanson, Wu, Zhang, Alizadeh, Zou (b55) 2023; 186
Tsamardinos, Aliferis (b60) 2003
Fourkioti, O., De Vries, M., Bakal, C., 2024. CAMIL: Context-Aware Multiple Instance Learning for Cancer Detection and Subtyping in Whole Slide Images. In: The Twelfth International Conference on Learning Representations. URL
Lu, Williamson, Chen, Chen, Barbieri, Mahmood (b42) 2021; 5
He, K., Zhang, X., Ren, S., Sun, J., 2016. Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. pp. 770–778.
Campanella, Hanna, Geneslaw, Miraflor, Werneck Krauss Silva, Busam, Brogi, Reuter, Klimstra, Fuchs (b5) 2019; 25
Cheng, Huang, Cai, Xu, Wang, Zhang (b11) 2025
Nasiri-Sarvi, A., Trinh, V.Q.-H., Rivaz, H., Hosseini, M.S., 2024. Vim4Path: Self-Supervised Vision Mamba for Histopathology Images. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 6894–6903.
Shi, Li, Gong, Fu (b50) 2024; 97
Fu, Yang, So, Lam, Bing, Collier (b19) 2023; 37
Kohavi, John (b33) 1997; 97
Shao, Bian, Chen, Wang, Zhang, Ji (b49) 2021; 34
Yang, Wang, Chen (b71) 2024
.
Ghaznavi, Evans, Madabhushi, Feldman (b21) 2013; 8
Wang, Yang, Zhang, Wang, Zhang, Yang, Huang, Han (b64) 2022; 81
Dosovitskiy (b16) 2020
Li, H., Zhu, C., Zhang, Y., Sun, Y., Shui, Z., Kuang, W., Zheng, S., Yang, L., 2023. Task-specific fine-tuning via variational information bottleneck for weakly-supervised pathology whole slide image classification. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 7454–7463.
Zheng, Gindra, Green, Burks, Betke, Beane, Kolachalama (b78) 2022; 41
Gadermayr, Tschuchnig (b20) 2024
Lee, J.C., Kwak, J.T., 2023. Order-ViT: Order Learning Vision Transformer for Cancer Classification in Pathology Images. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 2493–2502.
Srinidhi, Ciga, Martel (b52) 2021; 67
Bazargani, Fazli, Gleave, Goldenberg, Bashashati, Salcudean (b4) 2024; 96
Hendrycks, Gimpel (b25) 2016
Zhang, Li, Sun, Zheng, Zhu, Yang (b74) 2024
Tourniaire, Ilie, Hofman, Ayache, Delingette (b59) 2023; 85
Ehteshami Bejnordi, Veta, Johannes van Diest, van Ginneken, Karssemeijer, Litjens, van der Laak, the CAMELYON16 Consortium (b17) 2017; 318
Deng, Dong, Socher, Li, Li, Fei-Fei (b14) 2009
Lin, T., Xu, H., Yang, C., Xu, Y., 2022. Interventional multi-instance learning with deconfounded instance-level prediction. In: Proceedings of the AAAI Conference on Artificial Intelligence.
Carbonneau, Cheplygina, Granger, Gagnon (b6) 2018; 77
Li, Zhang, Chen, Shui, Zhu, Yang (b38) 2024
Xiang, Song, Zhang, Liu, Chen, Zhang, Huang, O’Donnell, Cai (b66) 2022; 41
Tang, W., Zhou, F., Huang, S., Zhu, X., Zhang, Y., Liu, B., 2024. Feature Re-Embedding: Towards Foundation Model-Level Performance in Computational Pathology. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 11343–11352.
Wang, Zhang, Su, Zhu (b65) 2024
Qu, Ma, Luo, Guo, Wang, Song (b47) 2024
DeLong, DeLong, Clarke-Pearson (b12) 1988
Hosseini, Bejnordi, Trinh, Chan, Hasan, Li, Yang, Kim, Zhang, Wu (b26) 2024
Bahadir, Omar, Rosenthal, Marchionni, Liechty, Pisapia, Sabuncu (b1) 2024
Li, J., Chen, Y., Chu, H., Sun, Q., Guan, T., Han, A., He, Y., 2024a. Dynamic Graph Representation with Knowledge-aware Attention for Histopathology Whole Slide Image Analysis. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 11323–11332.
Deng, Liu, Cui, Yao, Yue, Xiong, Yu, Wu, Yin, Wang (b15) 2024
Deng, Cui, Remedios, Bao, Womick, Chiron, Li, Roland, Lau, Liu (b13) 2022
Yang, Yan, Ming, Wang, Zhang, Tian (b72) 2021
Zhao, Deng, Li, Zhou, Gao, Wang, Li (b77) 2024; 94
Zhang, H., Meng, Y., Zhao, Y., Qiao, Y., Yang, X., Coupland, S.E., Zheng, Y., 2022a. DTFD-MIL: Double-tier feature distillation multiple instance learning for histopathology whole slide image classification. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 18802–18812.
Vaswani, Shazeer, Parmar, Uszkoreit, Jones, Gomez, Kaiser, Polosukhin (b62) 2017; 30
Li, B., Li, Y., Eliceiri, K.W., 2021b. Dual-stream multiple instance learning network for whole slide image classification with self-supervised contrastive learning. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 14318–14328.
Chen, Ding, Lu, Williamson, Jaume, Song, Chen, Zhang, Shao, Shaban (b8) 2024; 30
Li, Chen, Huang, Yang, Hu, Duan, Metaxas, Li, Zhang (b36) 2021
Yang, Liu, Ji (b70) 2024; 249
Van der Maaten, Hinton (b43) 2008; 9
Wang, Liu, Bocchieri, Li (b63) 2021; 34
Tang, W., Huang, S., Zhang, X., Zhou, F., Zhang, Y., Liu, B., 2023. Multiple instance learning framework with masked hard instance mining for whole slide image classification. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 4078–4087.
Zheng, Li, Shi, Xie, Huai, Cao, Jiang (b79) 2023; 42
Chen, Kornblith, Norouzi, Hinton (b9) 2020
Shi, Tang, Gao, Li, Wang, Gong, Li, Fu (b51) 2023; 42
Zhang, Zhu, Yang, Hosseini, Genovese, Chen, Rowsell, Damaskinos, Varma, Plataniotis (b76) 2022
Xu, Usuyama, Bagga, Zhang, Rao, Naumann, Wong, Gero, González, Gu (b67) 2024
Lin, Zhu, Cheng, Chen (b41) 2024
Xu, Zhu, Tang, Wang, Zhang, Li, Jiang, Shi, Liu, Jin (b69) 2021; 11
Houlsby, Giurgiu, Jastrzebski, Morrone, De Laroussilhe, Gesmundo, Attariyan, Gelly (b28) 2019
Kapse, S., Pati, P., Das, S., Zhang, J., Chen, C., Vakalopoulou, M., Saltz, J., Samaras, D., Gupta, R.R., Prasanna, P., 2024. SI-MIL: Taming Deep MIL for Self-Interpretability in Gigapixel Histopathology. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 11226–11237.
Zhao (10.1016/j.media.2025.103631_b77) 2024; 94
Xu (10.1016/j.media.2025.103631_b67) 2024
Grill (10.1016/j.media.2025.103631_b22) 2020; 33
Ilse (10.1016/j.media.2025.103631_b29) 2018
Bazargani (10.1016/j.media.2025.103631_b4) 2024; 96
Dosovitskiy (10.1016/j.media.2025.103631_b16) 2020
Li (10.1016/j.media.2025.103631_b36) 2021
10.1016/j.media.2025.103631_b40
Srinidhi (10.1016/j.media.2025.103631_b52) 2021; 67
Deng (10.1016/j.media.2025.103631_b13) 2022
10.1016/j.media.2025.103631_b45
10.1016/j.media.2025.103631_b48
Xu (10.1016/j.media.2025.103631_b69) 2021; 11
Houlsby (10.1016/j.media.2025.103631_b28) 2019
Zhang (10.1016/j.media.2025.103631_b76) 2022
Bandi (10.1016/j.media.2025.103631_b2) 2018; 38
Fu (10.1016/j.media.2025.103631_b19) 2023; 37
Swanson (10.1016/j.media.2025.103631_b55) 2023; 186
Shi (10.1016/j.media.2025.103631_b50) 2024; 97
Barbano (10.1016/j.media.2025.103631_b3) 2021
10.1016/j.media.2025.103631_b75
10.1016/j.media.2025.103631_b32
10.1016/j.media.2025.103631_b34
Ioffe (10.1016/j.media.2025.103631_b30) 2015
10.1016/j.media.2025.103631_b35
Chen (10.1016/j.media.2025.103631_b10) 2021
Shi (10.1016/j.media.2025.103631_b51) 2023; 42
10.1016/j.media.2025.103631_b37
10.1016/j.media.2025.103631_b39
Wang (10.1016/j.media.2025.103631_b64) 2022; 81
Yang (10.1016/j.media.2025.103631_b71) 2024
Jin (10.1016/j.media.2025.103631_b31) 2024
Deng (10.1016/j.media.2025.103631_b15) 2024
Ghaznavi (10.1016/j.media.2025.103631_b21) 2013; 8
Cheng (10.1016/j.media.2025.103631_b11) 2025
Ehteshami Bejnordi (10.1016/j.media.2025.103631_b17) 2017; 318
Tan (10.1016/j.media.2025.103631_b56) 2021
Zheng (10.1016/j.media.2025.103631_b78) 2022; 41
Li (10.1016/j.media.2025.103631_b38) 2024
Hosseini (10.1016/j.media.2025.103631_b26) 2024
Hendrycks (10.1016/j.media.2025.103631_b25) 2016
Lu (10.1016/j.media.2025.103631_b42) 2021; 5
Wang (10.1016/j.media.2025.103631_b63) 2021; 34
Yu (10.1016/j.media.2025.103631_b73) 2023; 85
Deng (10.1016/j.media.2025.103631_b14) 2009
Zhang (10.1016/j.media.2025.103631_b74) 2024
Lin (10.1016/j.media.2025.103631_b41) 2024
Chen (10.1016/j.media.2025.103631_b8) 2024; 30
Yang (10.1016/j.media.2025.103631_b70) 2024; 249
Yang (10.1016/j.media.2025.103631_b72) 2021
Paszke (10.1016/j.media.2025.103631_b46) 2019; 32
Tourniaire (10.1016/j.media.2025.103631_b59) 2023; 85
Wang (10.1016/j.media.2025.103631_b65) 2024
10.1016/j.media.2025.103631_b23
Vaswani (10.1016/j.media.2025.103631_b62) 2017; 30
10.1016/j.media.2025.103631_b24
Xu (10.1016/j.media.2025.103631_b68) 2025
10.1016/j.media.2025.103631_b27
Campanella (10.1016/j.media.2025.103631_b5) 2019; 25
DeLong (10.1016/j.media.2025.103631_b12) 1988
Carbonneau (10.1016/j.media.2025.103631_b6) 2018; 77
Kohavi (10.1016/j.media.2025.103631_b33) 1997; 97
Chen (10.1016/j.media.2025.103631_b9) 2020
Bahadir (10.1016/j.media.2025.103631_b1) 2024
10.1016/j.media.2025.103631_b7
Van der Maaten (10.1016/j.media.2025.103631_b43) 2008; 9
Qu (10.1016/j.media.2025.103631_b47) 2024
Tsamardinos (10.1016/j.media.2025.103631_b60) 2003
Marquette (10.1016/j.media.2025.103631_b44) 1995; 151
10.1016/j.media.2025.103631_b53
10.1016/j.media.2025.103631_b54
Shao (10.1016/j.media.2025.103631_b49) 2021; 34
Zheng (10.1016/j.media.2025.103631_b79) 2023; 42
10.1016/j.media.2025.103631_b57
10.1016/j.media.2025.103631_b58
Xiang (10.1016/j.media.2025.103631_b66) 2022; 41
10.1016/j.media.2025.103631_b18
Ulyanov (10.1016/j.media.2025.103631_b61) 2016
Gadermayr (10.1016/j.media.2025.103631_b20) 2024
References_xml – volume: 33
  start-page: 21271
  year: 2020
  end-page: 21284
  ident: b22
  article-title: Bootstrap your own latent-a new approach to self-supervised learning
  publication-title: Adv. Neural Inf. Process. Syst.
– reference: Tang, W., Huang, S., Zhang, X., Zhou, F., Zhang, Y., Liu, B., 2023. Multiple instance learning framework with masked hard instance mining for whole slide image classification. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 4078–4087.
– volume: 41
  start-page: 3003
  year: 2022
  end-page: 3015
  ident: b78
  article-title: A graph-transformer for whole slide image classification
  publication-title: IEEE Trans. Med. Imaging
– volume: 77
  start-page: 329
  year: 2018
  end-page: 353
  ident: b6
  article-title: Multiple instance learning: A survey of problem characteristics and applications
  publication-title: Pattern Recognit.
– reference: Hou, L., Samaras, D., Kurc, T.M., Gao, Y., Davis, J.E., Saltz, J.H., 2016. Patch-based convolutional neural network for whole slide tissue image classification. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. pp. 2424–2433.
– reference: Sultani, W., Chen, C., Shah, M., 2018. Real-world anomaly detection in surveillance videos. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. pp. 6479–6488.
– reference: He, X., Li, C., Zhang, P., Yang, J., Wang, X.E., 2023. Parameter-efficient model adaptation for vision transformers. In: Proceedings of the AAAI Conference on Artificial Intelligence. 37, (1), pp. 817–825.
– year: 2024
  ident: b38
  article-title: Rethinking transformer for long contextual histopathology whole slide image analysis
– volume: 186
  start-page: 1772
  year: 2023
  end-page: 1791
  ident: b55
  article-title: From patterns to patients: Advances in clinical machine learning for cancer diagnosis, prognosis, and treatment
  publication-title: Cell
– year: 2016
  ident: b61
  article-title: Instance normalization: The missing ingredient for fast stylization
– year: 2025
  ident: b68
  article-title: When multiple instance learning meets foundation models: advancing histological whole slide image analysis
  publication-title: Med. Image Anal.
– volume: 85
  year: 2023
  ident: b73
  article-title: Prototypical multiple instance learning for predicting lymph node metastasis of breast cancer from whole-slide pathological images
  publication-title: Med. Image Anal.
– reference: Kapse, S., Pati, P., Das, S., Zhang, J., Chen, C., Vakalopoulou, M., Saltz, J., Samaras, D., Gupta, R.R., Prasanna, P., 2024. SI-MIL: Taming Deep MIL for Self-Interpretability in Gigapixel Histopathology. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 11226–11237.
– reference: Ramon, J., 2000. Multi instance neural networks. In: ML-2000 Workshop Attribute-Value and Relational Learning.
– volume: 318
  start-page: 2199
  year: 2017
  end-page: 2210
  ident: b17
  article-title: Diagnostic assessment of deep learning algorithms for detection of lymph node metastases in women with breast cancer
  publication-title: JAMA
– volume: 34
  start-page: 29074
  year: 2021
  end-page: 29087
  ident: b63
  article-title: Can multi-label classification networks know what they don’t know?
  publication-title: Adv. Neural Inf. Process. Syst.
– reference: Li, B., Li, Y., Eliceiri, K.W., 2021b. Dual-stream multiple instance learning network for whole slide image classification with self-supervised contrastive learning. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 14318–14328.
– reference: Zhang, H., Meng, Y., Zhao, Y., Qiao, Y., Yang, X., Coupland, S.E., Zheng, Y., 2022a. DTFD-MIL: Double-tier feature distillation multiple instance learning for histopathology whole slide image classification. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 18802–18812.
– year: 2024
  ident: b26
  article-title: Computational pathology: a survey review and the way forward
  publication-title: J. Pathol. Informatics
– volume: 30
  start-page: 850
  year: 2024
  end-page: 862
  ident: b8
  article-title: Towards a general-purpose foundation model for computational pathology
  publication-title: Nature Med.
– reference: Lee, J.C., Kwak, J.T., 2023. Order-ViT: Order Learning Vision Transformer for Cancer Classification in Pathology Images. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 2493–2502.
– year: 2024
  ident: b47
  article-title: Rethinking multiple instance learning for whole slide image classification: A good instance classifier is all you need
  publication-title: IEEE Trans. Circuits Syst. Video Technol.
– start-page: 1597
  year: 2020
  end-page: 1607
  ident: b9
  article-title: A simple framework for contrastive learning of visual representations
  publication-title: International Conference on Machine Learning
– start-page: 339
  year: 2021
  end-page: 349
  ident: b10
  article-title: Whole slide images are 2d point clouds: Context-aware survival prediction using patch-based graph convolutional networks
  publication-title: Medical Image Computing and Computer Assisted Intervention–MICCAI 2021: 24th International Conference, Strasbourg, France, September 27–October 1, 2021, Proceedings, Part VIII 24
– reference: Sun, Q., Jiang, D., Li, J., Yan, R., He, Y., Guan, T., Cheng, Z., 2024. NcIEMIL: Rethinking Decoupled Multiple Instance Learning Framework for Histopathological Slide Classification. In: Medical Imaging with Deep Learning.
– volume: 8
  start-page: 331
  year: 2013
  end-page: 359
  ident: b21
  article-title: Digital imaging in pathology: whole-slide imaging and beyond
  publication-title: Annu. Rev. Pathol.: Mech. Dis.
– volume: 81
  year: 2022
  ident: b64
  article-title: Transformer-based unsupervised contrastive learning for histopathological image classification
  publication-title: Med. Image Anal.
– start-page: 309
  year: 2021
  end-page: 318
  ident: b36
  article-title: Hybrid supervision learning for pathology whole slide image classification
  publication-title: International Conference on Medical Image Computing and Computer-Assisted Intervention
– start-page: 24
  year: 2022
  end-page: 33
  ident: b13
  article-title: Cross-scale attention guided multi-instance learning for crohn’s disease diagnosis with pathological images
  publication-title: International Workshop on Multiscale Multimodal Medical Imaging
– reference: Li, H., Zhu, C., Zhang, Y., Sun, Y., Shui, Z., Kuang, W., Zheng, S., Yang, L., 2023. Task-specific fine-tuning via variational information bottleneck for weakly-supervised pathology whole slide image classification. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 7454–7463.
– volume: 42
  start-page: 3871
  year: 2023
  end-page: 3883
  ident: b51
  article-title: MG-trans: Multi-scale graph transformer with information bottleneck for whole slide image classification
  publication-title: IEEE Trans. Med. Imaging
– start-page: 837
  year: 1988
  end-page: 845
  ident: b12
  article-title: Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach
  publication-title: Biometrics
– start-page: 76
  year: 2021
  end-page: 80
  ident: b3
  article-title: Unitopatho, a labeled histopathological dataset for colorectal polyps classification and adenoma dysplasia grading
  publication-title: 2021 IEEE International Conference on Image Processing
– reference: Tang, W., Zhou, F., Huang, S., Zhu, X., Zhang, Y., Liu, B., 2024. Feature Re-Embedding: Towards Foundation Model-Level Performance in Computational Pathology. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 11343–11352.
– volume: 94
  year: 2024
  ident: b77
  article-title: LESS: Label-efficient multi-scale learning for cytological whole slide image screening
  publication-title: Med. Image Anal.
– reference: Chen, R.J., Chen, C., Li, Y., Chen, T.Y., Trister, A.D., Krishnan, R.G., Mahmood, F., 2022. Scaling vision transformers to gigapixel images via hierarchical self-supervised learning. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 16144–16155.
– volume: 25
  start-page: 1301
  year: 2019
  end-page: 1309
  ident: b5
  article-title: Clinical-grade computational pathology using weakly supervised deep learning on whole slide images
  publication-title: Nature Med.
– year: 2025
  ident: b11
  article-title: Focus your attention: Multiple instance learning with attention modification for whole slide pathological image classification
  publication-title: IEEE Trans. Circuits Syst. Video Technol.
– reference: Li, J., Chen, Y., Chu, H., Sun, Q., Guan, T., Han, A., He, Y., 2024a. Dynamic Graph Representation with Knowledge-aware Attention for Histopathology Whole Slide Image Analysis. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 11323–11332.
– year: 2016
  ident: b25
  article-title: Gaussian error linear units (gelus)
– year: 2024
  ident: b41
  article-title: Prompt-guided adaptive model transformation for whole slide image classification
– reference: Nasiri-Sarvi, A., Trinh, V.Q.-H., Rivaz, H., Hosseini, M.S., 2024. Vim4Path: Self-Supervised Vision Mamba for Histopathology Images. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 6894–6903.
– volume: 37
  start-page: 12799
  year: 2023
  end-page: 12807
  ident: b19
  article-title: On the effectiveness of parameter-efficient fine-tuning
  publication-title: Proceedings of the AAAI Conference on Artificial Intelligence
– volume: 30
  year: 2017
  ident: b62
  article-title: Attention is all you need
  publication-title: Adv. Neural Inf. Process. Syst.
– year: 2024
  ident: b71
  article-title: Mambamil: Enhancing long sequence modeling with sequence reordering in computational pathology
  publication-title: International Conference on Medical Image Computing and Computer-Assisted Intervention
– volume: 151
  start-page: 1878
  year: 1995
  end-page: 1888
  ident: b44
  article-title: Diagnostic tests for pneumonia in ventilated patients: prospective evaluation of diagnostic accuracy using histology as a diagnostic gold standard
  publication-title: Am. J. Respir. Crit. Care Med.
– year: 2015
  ident: b30
  article-title: Batch normalization: Accelerating deep network training by reducing internal covariate shift
– start-page: 1
  year: 2024
  end-page: 8
  ident: b67
  article-title: A whole-slide foundation model for digital pathology from real-world data
  publication-title: Nature
– year: 2024
  ident: b20
  article-title: Multiple instance learning for digital pathology: A review of the state-of-the-art, limitations & future potential
  publication-title: Comput. Med. Imaging Graph.
– year: 2024
  ident: b31
  article-title: HMIL: Hierarchical multi-instance learning for fine-grained whole slide image classification
  publication-title: IEEE Trans. Med. Imaging
– start-page: 2127
  year: 2018
  end-page: 2136
  ident: b29
  article-title: Attention-based deep multiple instance learning
  publication-title: International Conference on Machine Learning
– volume: 96
  year: 2024
  ident: b4
  article-title: Multi-scale relational graph convolutional network for multiple instance learning in histopathology images
  publication-title: Med. Image Anal.
– reference: He, K., Zhang, X., Ren, S., Sun, J., 2016. Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. pp. 770–778.
– volume: 249
  year: 2024
  ident: b70
  article-title: ProDiv: Prototype-driven consistent pseudo-bag division for whole-slide image classification
  publication-title: Comput. Methods Programs Biomed.
– volume: 67
  year: 2021
  ident: b52
  article-title: Deep neural network models for computational histopathology: A survey
  publication-title: Med. Image Anal.
– start-page: 300
  year: 2003
  end-page: 307
  ident: b60
  article-title: Towards principled feature selection: Relevancy, filters and wrappers
  publication-title: International Workshop on Artificial Intelligence and Statistics
– volume: 42
  start-page: 2726
  year: 2023
  end-page: 2739
  ident: b79
  article-title: Kernel attention transformer for histopathology whole slide image analysis and assistant cancer diagnosis
  publication-title: IEEE Trans. Med. Imaging
– start-page: 2790
  year: 2019
  end-page: 2799
  ident: b28
  article-title: Parameter-efficient transfer learning for NLP
  publication-title: International Conference on Machine Learning
– volume: 97
  start-page: 273
  year: 1997
  end-page: 324
  ident: b33
  article-title: Wrappers for feature subset selection
  publication-title: Artificial Intelligence
– start-page: 1276
  year: 2022
  end-page: 1280
  ident: b76
  article-title: Histokt: Cross knowledge transfer in computational pathology
  publication-title: ICASSP 2022-2022 IEEE International Conference on Acoustics, Speech and Signal Processing
– volume: 32
  year: 2019
  ident: b46
  article-title: Pytorch: An imperative style, high-performance deep learning library
  publication-title: Adv. Neural Inf. Process. Syst.
– volume: 85
  year: 2023
  ident: b59
  article-title: MS-CLAM: Mixed supervision for the classification and localization of tumors in whole slide images
  publication-title: Med. Image Anal.
– reference: Lin, T., Xu, H., Yang, C., Xu, Y., 2022. Interventional multi-instance learning with deconfounded instance-level prediction. In: Proceedings of the AAAI Conference on Artificial Intelligence.
– volume: 11
  year: 2021
  ident: b69
  article-title: Predicting axillary lymph node metastasis in early breast cancer using deep learning on primary tumor biopsy slides
  publication-title: Front. Oncol.
– start-page: 11830
  year: 2021
  end-page: 11841
  ident: b72
  article-title: Rethinking rotated object detection with gaussian wasserstein distance loss
  publication-title: International Conference on Machine Learning
– year: 2020
  ident: b16
  article-title: An image is worth 16×16 words: Transformers for image recognition at scale
– start-page: 11736
  year: 2024
  end-page: 11746
  ident: b15
  article-title: PrPSeg: Universal proposition learning for panoramic renal pathology segmentation
  publication-title: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition
– volume: 34
  start-page: 2136
  year: 2021
  end-page: 2147
  ident: b49
  article-title: Transmil: Transformer based correlated multiple instance learning for whole slide image classification
  publication-title: Adv. Neural Inf. Process. Syst.
– reference: Fourkioti, O., De Vries, M., Bakal, C., 2024. CAMIL: Context-Aware Multiple Instance Learning for Cancer Detection and Subtyping in Whole Slide Images. In: The Twelfth International Conference on Learning Representations. URL
– start-page: 248
  year: 2009
  end-page: 255
  ident: b14
  article-title: Imagenet: A large-scale hierarchical image database
  publication-title: 2009 IEEE Conference on Computer Vision and Pattern Recognition
– volume: 41
  start-page: 2180
  year: 2022
  end-page: 2190
  ident: b66
  article-title: Dsnet: A dual-stream framework for weakly-supervised gigapixel pathology image analysis
  publication-title: IEEE Trans. Med. Imaging
– reference: .
– volume: 38
  start-page: 550
  year: 2018
  end-page: 560
  ident: b2
  article-title: From detection of individual metastases to classification of lymph node status at the patient level: the camelyon17 challenge
  publication-title: IEEE Trans. Med. Imaging
– start-page: 1
  year: 2024
  end-page: 16
  ident: b1
  article-title: Artificial intelligence applications in histopathology
  publication-title: Nat. Rev. Electr. Eng.
– volume: 5
  start-page: 555
  year: 2021
  end-page: 570
  ident: b42
  article-title: Data-efficient and weakly supervised computational pathology on whole-slide images
  publication-title: Nat. Biomed. Eng.
– volume: 9
  year: 2008
  ident: b43
  article-title: Visualizing data using t-SNE
  publication-title: J. Mach. Learn. Res.
– year: 2024
  ident: b65
  article-title: A comprehensive survey of continual learning: Theory, method and application
  publication-title: IEEE Trans. Pattern Anal. Mach. Intell.
– year: 2024
  ident: b74
  article-title: Attention-challenging multiple instance learning for whole slide image classification
  publication-title: European Conference on Computer Vision
– volume: 97
  year: 2024
  ident: b50
  article-title: E2-MIL: An explainable and evidential multiple instance learning framework for whole slide image classification
  publication-title: Med. Image Anal.
– start-page: 10096
  year: 2021
  end-page: 10106
  ident: b56
  article-title: Efficientnetv2: Smaller models and faster training
  publication-title: International Conference on Machine Learning
– year: 2025
  ident: 10.1016/j.media.2025.103631_b11
  article-title: Focus your attention: Multiple instance learning with attention modification for whole slide pathological image classification
  publication-title: IEEE Trans. Circuits Syst. Video Technol.
  doi: 10.1109/TCSVT.2025.3528625
– volume: 85
  year: 2023
  ident: 10.1016/j.media.2025.103631_b73
  article-title: Prototypical multiple instance learning for predicting lymph node metastasis of breast cancer from whole-slide pathological images
  publication-title: Med. Image Anal.
  doi: 10.1016/j.media.2023.102748
– volume: 33
  start-page: 21271
  year: 2020
  ident: 10.1016/j.media.2025.103631_b22
  article-title: Bootstrap your own latent-a new approach to self-supervised learning
  publication-title: Adv. Neural Inf. Process. Syst.
– ident: 10.1016/j.media.2025.103631_b58
  doi: 10.1109/CVPR52733.2024.01078
– year: 2024
  ident: 10.1016/j.media.2025.103631_b74
  article-title: Attention-challenging multiple instance learning for whole slide image classification
– ident: 10.1016/j.media.2025.103631_b27
  doi: 10.1109/CVPR.2016.266
– volume: 34
  start-page: 2136
  year: 2021
  ident: 10.1016/j.media.2025.103631_b49
  article-title: Transmil: Transformer based correlated multiple instance learning for whole slide image classification
  publication-title: Adv. Neural Inf. Process. Syst.
– volume: 30
  year: 2017
  ident: 10.1016/j.media.2025.103631_b62
  article-title: Attention is all you need
  publication-title: Adv. Neural Inf. Process. Syst.
– volume: 96
  year: 2024
  ident: 10.1016/j.media.2025.103631_b4
  article-title: Multi-scale relational graph convolutional network for multiple instance learning in histopathology images
  publication-title: Med. Image Anal.
  doi: 10.1016/j.media.2024.103197
– start-page: 10096
  year: 2021
  ident: 10.1016/j.media.2025.103631_b56
  article-title: Efficientnetv2: Smaller models and faster training
– start-page: 1
  year: 2024
  ident: 10.1016/j.media.2025.103631_b67
  article-title: A whole-slide foundation model for digital pathology from real-world data
  publication-title: Nature
– start-page: 2790
  year: 2019
  ident: 10.1016/j.media.2025.103631_b28
  article-title: Parameter-efficient transfer learning for NLP
– volume: 5
  start-page: 555
  issue: 6
  year: 2021
  ident: 10.1016/j.media.2025.103631_b42
  article-title: Data-efficient and weakly supervised computational pathology on whole-slide images
  publication-title: Nat. Biomed. Eng.
  doi: 10.1038/s41551-020-00682-w
– volume: 81
  year: 2022
  ident: 10.1016/j.media.2025.103631_b64
  article-title: Transformer-based unsupervised contrastive learning for histopathological image classification
  publication-title: Med. Image Anal.
  doi: 10.1016/j.media.2022.102559
– start-page: 300
  year: 2003
  ident: 10.1016/j.media.2025.103631_b60
  article-title: Towards principled feature selection: Relevancy, filters and wrappers
– year: 2016
  ident: 10.1016/j.media.2025.103631_b61
– ident: 10.1016/j.media.2025.103631_b18
– ident: 10.1016/j.media.2025.103631_b35
  doi: 10.1109/CVPR52733.2024.01076
– volume: 85
  year: 2023
  ident: 10.1016/j.media.2025.103631_b59
  article-title: MS-CLAM: Mixed supervision for the classification and localization of tumors in whole slide images
  publication-title: Med. Image Anal.
  doi: 10.1016/j.media.2023.102763
– volume: 151
  start-page: 1878
  issue: 6
  year: 1995
  ident: 10.1016/j.media.2025.103631_b44
  article-title: Diagnostic tests for pneumonia in ventilated patients: prospective evaluation of diagnostic accuracy using histology as a diagnostic gold standard
  publication-title: Am. J. Respir. Crit. Care Med.
  doi: 10.1164/ajrccm.151.6.7767535
– ident: 10.1016/j.media.2025.103631_b75
  doi: 10.1109/CVPR52688.2022.01824
– volume: 77
  start-page: 329
  year: 2018
  ident: 10.1016/j.media.2025.103631_b6
  article-title: Multiple instance learning: A survey of problem characteristics and applications
  publication-title: Pattern Recognit.
  doi: 10.1016/j.patcog.2017.10.009
– start-page: 339
  year: 2021
  ident: 10.1016/j.media.2025.103631_b10
  article-title: Whole slide images are 2d point clouds: Context-aware survival prediction using patch-based graph convolutional networks
– volume: 38
  start-page: 550
  issue: 2
  year: 2018
  ident: 10.1016/j.media.2025.103631_b2
  article-title: From detection of individual metastases to classification of lymph node status at the patient level: the camelyon17 challenge
  publication-title: IEEE Trans. Med. Imaging
  doi: 10.1109/TMI.2018.2867350
– ident: 10.1016/j.media.2025.103631_b39
  doi: 10.1109/CVPR52729.2023.00720
– ident: 10.1016/j.media.2025.103631_b53
  doi: 10.1109/CVPR.2018.00678
– volume: 186
  start-page: 1772
  issue: 8
  year: 2023
  ident: 10.1016/j.media.2025.103631_b55
  article-title: From patterns to patients: Advances in clinical machine learning for cancer diagnosis, prognosis, and treatment
  publication-title: Cell
  doi: 10.1016/j.cell.2023.01.035
– ident: 10.1016/j.media.2025.103631_b23
  doi: 10.1609/aaai.v37i1.25160
– ident: 10.1016/j.media.2025.103631_b34
  doi: 10.1109/ICCVW60793.2023.00263
– start-page: 24
  year: 2022
  ident: 10.1016/j.media.2025.103631_b13
  article-title: Cross-scale attention guided multi-instance learning for crohn’s disease diagnosis with pathological images
– start-page: 2127
  year: 2018
  ident: 10.1016/j.media.2025.103631_b29
  article-title: Attention-based deep multiple instance learning
– ident: 10.1016/j.media.2025.103631_b48
– start-page: 11830
  year: 2021
  ident: 10.1016/j.media.2025.103631_b72
  article-title: Rethinking rotated object detection with gaussian wasserstein distance loss
– year: 2024
  ident: 10.1016/j.media.2025.103631_b47
  article-title: Rethinking multiple instance learning for whole slide image classification: A good instance classifier is all you need
  publication-title: IEEE Trans. Circuits Syst. Video Technol.
  doi: 10.1109/TCSVT.2024.3400876
– ident: 10.1016/j.media.2025.103631_b45
  doi: 10.1109/CVPRW63382.2024.00683
– volume: 249
  year: 2024
  ident: 10.1016/j.media.2025.103631_b70
  article-title: ProDiv: Prototype-driven consistent pseudo-bag division for whole-slide image classification
  publication-title: Comput. Methods Programs Biomed.
  doi: 10.1016/j.cmpb.2024.108161
– year: 2020
  ident: 10.1016/j.media.2025.103631_b16
– volume: 11
  year: 2021
  ident: 10.1016/j.media.2025.103631_b69
  article-title: Predicting axillary lymph node metastasis in early breast cancer using deep learning on primary tumor biopsy slides
  publication-title: Front. Oncol.
  doi: 10.3389/fonc.2021.759007
– start-page: 76
  year: 2021
  ident: 10.1016/j.media.2025.103631_b3
  article-title: Unitopatho, a labeled histopathological dataset for colorectal polyps classification and adenoma dysplasia grading
– start-page: 837
  year: 1988
  ident: 10.1016/j.media.2025.103631_b12
  article-title: Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach
  publication-title: Biometrics
  doi: 10.2307/2531595
– volume: 9
  issue: 11
  year: 2008
  ident: 10.1016/j.media.2025.103631_b43
  article-title: Visualizing data using t-SNE
  publication-title: J. Mach. Learn. Res.
– start-page: 248
  year: 2009
  ident: 10.1016/j.media.2025.103631_b14
  article-title: Imagenet: A large-scale hierarchical image database
– ident: 10.1016/j.media.2025.103631_b40
  doi: 10.1609/aaai.v36i2.20051
– volume: 8
  start-page: 331
  year: 2013
  ident: 10.1016/j.media.2025.103631_b21
  article-title: Digital imaging in pathology: whole-slide imaging and beyond
  publication-title: Annu. Rev. Pathol.: Mech. Dis.
  doi: 10.1146/annurev-pathol-011811-120902
– year: 2024
  ident: 10.1016/j.media.2025.103631_b65
  article-title: A comprehensive survey of continual learning: Theory, method and application
  publication-title: IEEE Trans. Pattern Anal. Mach. Intell.
– volume: 318
  start-page: 2199
  issue: 22
  year: 2017
  ident: 10.1016/j.media.2025.103631_b17
  article-title: Diagnostic assessment of deep learning algorithms for detection of lymph node metastases in women with breast cancer
  publication-title: JAMA
  doi: 10.1001/jama.2017.14585
– year: 2024
  ident: 10.1016/j.media.2025.103631_b38
– volume: 42
  start-page: 2726
  issue: 9
  year: 2023
  ident: 10.1016/j.media.2025.103631_b79
  article-title: Kernel attention transformer for histopathology whole slide image analysis and assistant cancer diagnosis
  publication-title: IEEE Trans. Med. Imaging
  doi: 10.1109/TMI.2023.3264781
– volume: 42
  start-page: 3871
  issue: 12
  year: 2023
  ident: 10.1016/j.media.2025.103631_b51
  article-title: MG-trans: Multi-scale graph transformer with information bottleneck for whole slide image classification
  publication-title: IEEE Trans. Med. Imaging
  doi: 10.1109/TMI.2023.3313252
– start-page: 1276
  year: 2022
  ident: 10.1016/j.media.2025.103631_b76
  article-title: Histokt: Cross knowledge transfer in computational pathology
– year: 2024
  ident: 10.1016/j.media.2025.103631_b26
  article-title: Computational pathology: a survey review and the way forward
  publication-title: J. Pathol. Informatics
  doi: 10.1016/j.jpi.2023.100357
– ident: 10.1016/j.media.2025.103631_b32
  doi: 10.1109/CVPR52733.2024.01067
– ident: 10.1016/j.media.2025.103631_b57
  doi: 10.1109/ICCV51070.2023.00377
– ident: 10.1016/j.media.2025.103631_b7
  doi: 10.1109/CVPR52688.2022.01567
– volume: 37
  start-page: 12799
  year: 2023
  ident: 10.1016/j.media.2025.103631_b19
  article-title: On the effectiveness of parameter-efficient fine-tuning
– ident: 10.1016/j.media.2025.103631_b24
  doi: 10.1109/CVPR.2016.90
– ident: 10.1016/j.media.2025.103631_b54
– start-page: 1597
  year: 2020
  ident: 10.1016/j.media.2025.103631_b9
  article-title: A simple framework for contrastive learning of visual representations
– year: 2016
  ident: 10.1016/j.media.2025.103631_b25
– year: 2024
  ident: 10.1016/j.media.2025.103631_b31
  article-title: HMIL: Hierarchical multi-instance learning for fine-grained whole slide image classification
  publication-title: IEEE Trans. Med. Imaging
– volume: 94
  year: 2024
  ident: 10.1016/j.media.2025.103631_b77
  article-title: LESS: Label-efficient multi-scale learning for cytological whole slide image screening
  publication-title: Med. Image Anal.
  doi: 10.1016/j.media.2024.103109
– year: 2015
  ident: 10.1016/j.media.2025.103631_b30
– year: 2025
  ident: 10.1016/j.media.2025.103631_b68
  article-title: When multiple instance learning meets foundation models: advancing histological whole slide image analysis
  publication-title: Med. Image Anal.
  doi: 10.1016/j.media.2025.103456
– volume: 97
  year: 2024
  ident: 10.1016/j.media.2025.103631_b50
  article-title: E2-MIL: An explainable and evidential multiple instance learning framework for whole slide image classification
  publication-title: Med. Image Anal.
  doi: 10.1016/j.media.2024.103294
– volume: 34
  start-page: 29074
  year: 2021
  ident: 10.1016/j.media.2025.103631_b63
  article-title: Can multi-label classification networks know what they don’t know?
  publication-title: Adv. Neural Inf. Process. Syst.
– ident: 10.1016/j.media.2025.103631_b37
  doi: 10.1109/CVPR46437.2021.01409
– start-page: 1
  year: 2024
  ident: 10.1016/j.media.2025.103631_b1
  article-title: Artificial intelligence applications in histopathology
  publication-title: Nat. Rev. Electr. Eng.
– start-page: 309
  year: 2021
  ident: 10.1016/j.media.2025.103631_b36
  article-title: Hybrid supervision learning for pathology whole slide image classification
– volume: 32
  year: 2019
  ident: 10.1016/j.media.2025.103631_b46
  article-title: Pytorch: An imperative style, high-performance deep learning library
  publication-title: Adv. Neural Inf. Process. Syst.
– volume: 41
  start-page: 2180
  issue: 8
  year: 2022
  ident: 10.1016/j.media.2025.103631_b66
  article-title: Dsnet: A dual-stream framework for weakly-supervised gigapixel pathology image analysis
  publication-title: IEEE Trans. Med. Imaging
  doi: 10.1109/TMI.2022.3157983
– volume: 41
  start-page: 3003
  issue: 11
  year: 2022
  ident: 10.1016/j.media.2025.103631_b78
  article-title: A graph-transformer for whole slide image classification
  publication-title: IEEE Trans. Med. Imaging
  doi: 10.1109/TMI.2022.3176598
– year: 2024
  ident: 10.1016/j.media.2025.103631_b71
  article-title: Mambamil: Enhancing long sequence modeling with sequence reordering in computational pathology
– year: 2024
  ident: 10.1016/j.media.2025.103631_b20
  article-title: Multiple instance learning for digital pathology: A review of the state-of-the-art, limitations & future potential
  publication-title: Comput. Med. Imaging Graph.
  doi: 10.1016/j.compmedimag.2024.102337
– volume: 97
  start-page: 273
  issue: 1–2
  year: 1997
  ident: 10.1016/j.media.2025.103631_b33
  article-title: Wrappers for feature subset selection
  publication-title: Artificial Intelligence
  doi: 10.1016/S0004-3702(97)00043-X
– volume: 25
  start-page: 1301
  issue: 8
  year: 2019
  ident: 10.1016/j.media.2025.103631_b5
  article-title: Clinical-grade computational pathology using weakly supervised deep learning on whole slide images
  publication-title: Nature Med.
  doi: 10.1038/s41591-019-0508-1
– volume: 30
  start-page: 850
  issue: 3
  year: 2024
  ident: 10.1016/j.media.2025.103631_b8
  article-title: Towards a general-purpose foundation model for computational pathology
  publication-title: Nature Med.
  doi: 10.1038/s41591-024-02857-3
– start-page: 11736
  year: 2024
  ident: 10.1016/j.media.2025.103631_b15
  article-title: PrPSeg: Universal proposition learning for panoramic renal pathology segmentation
– volume: 67
  year: 2021
  ident: 10.1016/j.media.2025.103631_b52
  article-title: Deep neural network models for computational histopathology: A survey
  publication-title: Med. Image Anal.
  doi: 10.1016/j.media.2020.101813
– year: 2024
  ident: 10.1016/j.media.2025.103631_b41
SSID ssj0007440
Score 2.4666317
Snippet Multiple instance learning (MIL) is a powerful approach for whole-slide pathological image (WSI) analysis, particularly suited for processing...
SourceID proquest
pubmed
crossref
elsevier
SourceType Aggregation Database
Index Database
Enrichment Source
Publisher
StartPage 103631
SubjectTerms Algorithms
Attribute scoring mechanism
Humans
Image Interpretation, Computer-Assisted - methods
Machine Learning
Multiple instance learning
Multiple-Instance Learning Algorithms
Pathological image analysis
Pathology adaptive learning
Pathology attribute constraint
Title AttriMIL: Revisiting attention-based multiple instance learning for whole-slide pathological image classification from a perspective of instance attributes
URI https://dx.doi.org/10.1016/j.media.2025.103631
https://www.ncbi.nlm.nih.gov/pubmed/40381256
https://www.proquest.com/docview/3205662470
Volume 103
WOSCitedRecordID wos001495110600001&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/eLvHCXMwtV1bb9MwFLa6Dk3jAcG4lctkJB5JlTgXO7xVaGhD28TDkMpT5DgOS9WlVZeO_Rf-AD-T41sSNjrBAy9W5SQnlc6Xc7G_c4zQ2zCQhHHCPRKL1IsEKTxwQ8KLilLVUYaslEwfNkFPT9l0mn4eDH66WpirOa1rdn2dLv-rqmEOlK1KZ_9B3a1QmIDfoHQYQe0w_pXiJ02zqk6Ojg3VTdWOa2az6qOpmY2eclxFxySsdIAIn_fcrZIo5uF3dW6uB0FoIVXv1c5GVheK5SNU0K1YRparqIpUuOqB7Co3TSsKK5k35lwty1d0B0jZPSIjkdvuKN2mSGXXDL6ddwg-XLv17XNZz6p2vl33_toC9XitEVrVS2m9s13cIHFLhAXfZAxymAQei0xNcmux_bBncwO1Fx380R2YlYnZWFfhjNULxrfvBnUsLzQYIrVtSuIbrbm1s3eXttA2oXHKhmh7cnQw_dR6fdVo0XW10vzBW-_cRTtOyqYgaFOSo4Ods4fogc1S8MSg6xEayHoP3e_1rtxDOyeWlfEY_XCQe487wOEbgMMOcNjBAjvAYQAc7gEO9wGHNTzw74DDCnCY4x7g8KLsJHeAe4K-fDw4-3Do2UM_PAGxZeOlVMggZDSnPi2FJDkLQy5ICnkw5BYkL6RkJfglyUSUBHlAS8hQEhEFuR9yMDrhUzSsF7V8jnBc0gIERT5nMJQBDwRNiYxzIYhMi2SEiNNCJmxHfHUwyzxz1MdZprWYKS1mRosj9K59aGkawtx9e-LUm9mY1sSqGWD07gffODBkYPHVNh6v5WJ9mYUEkpaERNQfoWcGJe0_cQB7sfHKS7TbfWiv0LBZreVrdE9cNdXlah9t0Snbt9j-BcBG18U
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=AttriMIL%3A+Revisiting+attention-based+multiple+instance+learning+for+whole-slide+pathological+image+classification+from+a+perspective+of+instance+attributes&rft.jtitle=Medical+image+analysis&rft.au=Cai%2C+Linghan&rft.au=Huang%2C+Shenjin&rft.au=Zhang%2C+Ye&rft.au=Lu%2C+Jinpeng&rft.date=2025-07-01&rft.eissn=1361-8423&rft.volume=103&rft.spage=103631&rft_id=info:doi/10.1016%2Fj.media.2025.103631&rft_id=info%3Apmid%2F40381256&rft.externalDocID=40381256
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