Disentangled Pseudo-Bag Augmentation for Whole Slide Image Multiple Instance Learning

As the predominant approach for pathological whole slide image (WSI) classification, multiple instance learning (MIL) methods struggle with limited labeled WSIs. Although MIL has achieved notable progress with pseudo-bag-oriented augmentation methods, their effectiveness is often constrained by nois...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:IEEE transactions on medical imaging Ročník 44; číslo 11; s. 4181 - 4197
Hlavní autoři: Dong, Jiuyang, Jiang, Junjun, Jiang, Kui, Li, Jiahan, Cai, Linghan, Zhang, Yongbing
Médium: Journal Article
Jazyk:angličtina
Vydáno: United States IEEE 01.11.2025
Témata:
Accuracy
Algorithms
Cancer
Data augmentation
Data mining
Feature extraction
Humans
Image Interpretation, Computer-Assisted - methods
Machine Learning
multiple instance learning
Multiple-Instance Learning Algorithms
Neural Networks, Computer
Pathology
Phenotypes
Probability
Training
Tumors
Whole slide image classification
ISSN:0278-0062, 1558-254X, 1558-254X
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:As the predominant approach for pathological whole slide image (WSI) classification, multiple instance learning (MIL) methods struggle with limited labeled WSIs. Although MIL has achieved notable progress with pseudo-bag-oriented augmentation methods, their effectiveness is often constrained by noisy pseudo-labels and low-quality pseudo-bags. To overcome these problems, we revisit the use of pseudo-bags for WSI data augmentation and propose a new pseudo-bag generation paradigm, dubbed DPBAug. Its distinctive features can be summarized as: i) We develop an intra-slide pseudo-bag generation module, which separates the heterogeneous instances within each slide through phenotype partitioning. Moreover, to ensure accurate label inheritance when generating pseudo-bags, we propose an instance sampling algorithm with replacement. ii) An inter-slide pseudo-bag fusion module is designed to integrate heterogeneous information across multiple WSIs, producing high-quality training samples that better leverage the potential of neural networks. iii) A pseudo-bag memory update module prioritizes valuable synthetic pseudo-bags. This further enhances the network's classification performance. Extensive experiments demonstrate that DPBAug surpasses existing augmentation methods, enhancing the classification performance and reliability of multiple MIL baselines across various public datasets. DPBAug also improves the generalization and data efficiency of existing MIL methods, facilitating their adoption in clinical practice and rare cancer research The project is available at: https://github.com/JiuyangDong/DPBAug .
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0278-0062
1558-254X
1558-254X
DOI:10.1109/TMI.2025.3569941