Dynamic Graph Representation for WSI Classification: A Knowledge-Aware Attention Mechanism for Enhanced Computational Pathology

Computational pathologists are now concentrating on developing methods for histopathological WSI analysis using deep learning. Methods using Transformer as a foundation are widely debated in the present paradigm, which is largely built on multiple instances learning (MIL). By using WSI sequence toke...

Full description

Saved in:
Bibliographic Details
Published in:2025 International Conference on Cognitive Computing in Engineering, Communications, Sciences and Biomedical Health Informatics (IC3ECSBHI) pp. 1084 - 1088
Main Authors: Kumar, T. Rajesh, R, Mahaveerakannan, R., Balamanigandan, Subramanian, P., Mallampalli, Praveena
Format: Conference Proceeding
Language:English
Published: IEEE 16.01.2025
Subjects:
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract Computational pathologists are now concentrating on developing methods for histopathological WSI analysis using deep learning. Methods using Transformer as a foundation are widely debated in the present paradigm, which is largely built on multiple instances learning (MIL). By using WSI sequence tokens to represent patches; these methods transform WSI tasks into sequence tasks. Gigapixel scale and high heterogeneity both contribute to feature complexity, which in turn causes Transformer-based MIL to struggle with issues including sluggish inference speed, large memory usage, and poor performance. Furthermore, the intricate structural relationships bet to en biological entities (such as the varied interactions bet to en various cell types) in the WSI cannot be mined using graph-based approaches. The research suggests a solution by outlining an ideal algorithm for dynamic graph representation that views WSIs as a subset of the knowledge graph structure. In particular, the research uses the head-tail interactions bet to en instances to build directed edge embeddings and neighbours dynamically. Afterwards, to come up with a knowledge-aware attention mechanism that can refresh the features of the head node by acquiring the combined attention score of every neighbour and edge. At last, the Puma optimiser algorithm performs the fine-tuning, and thanks to the updated head's global pooling process, to have an embedding at the graph level that can be used as an implicit illustration for WSI classification. Our system significantly outperforms state-of-the-art tactics on multiple tasks, as shown by extensive testing on three public TCGA benchmark datasets.
AbstractList Computational pathologists are now concentrating on developing methods for histopathological WSI analysis using deep learning. Methods using Transformer as a foundation are widely debated in the present paradigm, which is largely built on multiple instances learning (MIL). By using WSI sequence tokens to represent patches; these methods transform WSI tasks into sequence tasks. Gigapixel scale and high heterogeneity both contribute to feature complexity, which in turn causes Transformer-based MIL to struggle with issues including sluggish inference speed, large memory usage, and poor performance. Furthermore, the intricate structural relationships bet to en biological entities (such as the varied interactions bet to en various cell types) in the WSI cannot be mined using graph-based approaches. The research suggests a solution by outlining an ideal algorithm for dynamic graph representation that views WSIs as a subset of the knowledge graph structure. In particular, the research uses the head-tail interactions bet to en instances to build directed edge embeddings and neighbours dynamically. Afterwards, to come up with a knowledge-aware attention mechanism that can refresh the features of the head node by acquiring the combined attention score of every neighbour and edge. At last, the Puma optimiser algorithm performs the fine-tuning, and thanks to the updated head's global pooling process, to have an embedding at the graph level that can be used as an implicit illustration for WSI classification. Our system significantly outperforms state-of-the-art tactics on multiple tasks, as shown by extensive testing on three public TCGA benchmark datasets.
Author Subramanian, P.
Mallampalli, Praveena
Kumar, T. Rajesh
R., Balamanigandan
R, Mahaveerakannan
Author_xml – sequence: 1
  givenname: T. Rajesh
  surname: Kumar
  fullname: Kumar, T. Rajesh
  email: t.rajesh61074@gmail.com
  organization: Saveetha School of Engineering, Saveetha Institute of Medical and Technical Sciences,Department of Computer Science and Engineering,Chennai,Tamil Nadu,India,602105
– sequence: 2
  givenname: Mahaveerakannan
  surname: R
  fullname: R, Mahaveerakannan
  email: mahaveerakannanr.sse@saveetha.com
  organization: Saveetha School of Engineering, Saveetha Institute of Medical and Technical Sciences,Department of Computer Science and Engineering,Chennai,Tamil Nadu,India,602105
– sequence: 3
  givenname: Balamanigandan
  surname: R.
  fullname: R., Balamanigandan
  email: balamanigandanr.sse@saveetha.com
  organization: Saveetha School of Engineering, Saveetha Institute of Medical and Technical Sciences,Department of Computer Science and Engineering,Chennai,Tamil Nadu,India,602105
– sequence: 4
  givenname: P.
  surname: Subramanian
  fullname: Subramanian, P.
  email: subramanianp.sse@saveetha.com
  organization: Saveetha School of Engineering, Saveetha Institute of Medical and Technical Sciences,Department of Networking, Department of Computer Science and Engineering,Chennai,Tamil Nadu,India
– sequence: 5
  givenname: Praveena
  surname: Mallampalli
  fullname: Mallampalli, Praveena
  email: praveena.iduri@gmail.com
  organization: Swarnandhra College of Engineering and Technology,Department of Computer Science,Narsapur,Andhra Pradesh,India
BookMark eNo1kD1PwzAURY0EA5T-AwYPrCm1HbsxWwj9iCgC0UqM1XPy3FpK7CgJqjrx1yktTFf3SucM94Zc-uCRkHs2HjE21g95JqbZ6mmRKyE1G_Exl6Pjrhnn-oIM9UQnQjDJEsnFNfl-PnioXUHnLTQ7-oFNix36HnoXPLWhpZ-rnGYVdJ2zrjjNjzSlLz7sKyy3GKV7aJGmfX-kfplXLHbgXVef6Kk_lgJLmoW6-TproaLv0O9CFbaHW3Jloepw-JcDsp5N19kiWr7N8yxdRk6LPrKy1AqVFMyawiQA1pTaxBJsgbFhwnKFcWK01GhRTyZGG8MTVKBiDirhYkDuzlqHiJumdTW0h83_LeIH5VFhow
ContentType Conference Proceeding
DBID 6IE
6IL
CBEJK
RIE
RIL
DOI 10.1109/IC3ECSBHI63591.2025.10991229
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Xplore POP ALL
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP All) 1998-Present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
EISBN 9798331518523
EndPage 1088
ExternalDocumentID 10991229
Genre orig-research
GroupedDBID 6IE
6IL
CBEJK
RIE
RIL
ID FETCH-LOGICAL-i93t-f5d96e6531fbcb8aafbd9b45afce4b13f26e48b959efe977b9bb28e6a642a6823
IEDL.DBID RIE
IngestDate Thu May 29 05:57:30 EDT 2025
IsPeerReviewed false
IsScholarly false
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i93t-f5d96e6531fbcb8aafbd9b45afce4b13f26e48b959efe977b9bb28e6a642a6823
PageCount 5
ParticipantIDs ieee_primary_10991229
PublicationCentury 2000
PublicationDate 2025-Jan.-16
PublicationDateYYYYMMDD 2025-01-16
PublicationDate_xml – month: 01
  year: 2025
  text: 2025-Jan.-16
  day: 16
PublicationDecade 2020
PublicationTitle 2025 International Conference on Cognitive Computing in Engineering, Communications, Sciences and Biomedical Health Informatics (IC3ECSBHI)
PublicationTitleAbbrev IC3ECSBHI
PublicationYear 2025
Publisher IEEE
Publisher_xml – name: IEEE
Score 1.8960891
Snippet Computational pathologists are now concentrating on developing methods for histopathological WSI analysis using deep learning. Methods using Transformer as a...
SourceID ieee
SourceType Publisher
StartPage 1084
SubjectTerms Attention mechanisms
Classification algorithms
Deep learning
Graph-based methods
Head
Heuristic algorithms
Histopathological whole slide image
Inference algorithms
Knowledge graph structure
Knowledge graphs
Multiple instance learning
Optimal dynamic graph representation
Pathology
Puma optimizer algorithm
Transformers
Transforms
Title Dynamic Graph Representation for WSI Classification: A Knowledge-Aware Attention Mechanism for Enhanced Computational Pathology
URI https://ieeexplore.ieee.org/document/10991229
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1NTwIxEG2EGONJjRi_04PXAtvtLh1viCDESIiQyI20u9PIwcXAovHkX7ftLhoPHrw1Tdombdr56HvzCLkKUImmUSmTAjRzAEamZWQYCK5NYGOuFBMvNtEaDuV0CqOSrO65MIjowWdYd03_l58ukrVLlTXcL07AOVRIpdVqFWStHbteUTezMeiE3c74pj-wRhRc7Mej-mbIL_EUbzt6e_9cdZ_Uflh4dPRtXw7IFmaH5PO2kJCnd67SNH30QNaSP5RR64HSp_GAeqlLBwLy3de0Te83uTPWfldLpO08L4CO9AEd-Xe-evGju9mzxwTQQu6hTBXSkcr9M_lRI5Ned9Lps1JEgc0hzJmJUogxtjfN6ERLpYxOQYtImQSFDkLDYxRSQwRo0PqCGrTmEmNl4xIVSx4ekWq2yPCYUGgKO0mk7BOQ2CDSALfOE0ghuQ4hTJMTUnObN3stymTMNvt2-kf_Gdl1R-TyGUF8Tqr5co0XZDt5y-er5aU_3C_Ay6nA
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1NT8MgGCY6jXpS44zfcvDKtlJawducm2v2kcUtcbcF2pdsBzuzdRpP_nWBdhoPHrwREiCBwPvB87wPQjceSFbTMiGcCUUsgJEoHmgiGFXaMzFXArETm7jt9_l4LAYFWd1xYQDAgc-gYpvuLz-ZxyubKqvaXxyPUrGJtgLGqJfTtXbMinnlzGrU8JuN4X07MmZU2OiPBpX1oF_yKc56tPb_ue4BKv_w8PDg28Icog1Ij9DnQy4ijx9trWn85KCsBYMoxcYHxc_DCDuxSwsDct13uI476-wZqb_LBeB6luVQR9wDS_-dLV_c6GY6dagAnAs-FMlCPJCZeyg_ymjUao4abVLIKJCZ8DOig0SEEJq7plWsuJRaJUKxQOoYmPJ8TUNgXIlAgAbjDSqhFOUQShOZyJBT_xiV0nkKJwiLGjOTBNI8ArEJI7Wgxn0SnHGqfOEn8Skq282bvOaFMibrfTv7o_8a7bZHve6kG_U752jPHpfNbnjhBSplixVcou34LZstF1fuoL8AvpGtBw
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=proceeding&rft.title=2025+International+Conference+on+Cognitive+Computing+in+Engineering%2C+Communications%2C+Sciences+and+Biomedical+Health+Informatics+%28IC3ECSBHI%29&rft.atitle=Dynamic+Graph+Representation+for+WSI+Classification%3A+A+Knowledge-Aware+Attention+Mechanism+for+Enhanced+Computational+Pathology&rft.au=Kumar%2C+T.+Rajesh&rft.au=R%2C+Mahaveerakannan&rft.au=R.%2C+Balamanigandan&rft.au=Subramanian%2C+P.&rft.date=2025-01-16&rft.pub=IEEE&rft.spage=1084&rft.epage=1088&rft_id=info:doi/10.1109%2FIC3ECSBHI63591.2025.10991229&rft.externalDocID=10991229