EUVIMEDEUVIMED entdecken
Crossref

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

Jiuyang Dong, Junjun Jiang, Kui Jiang, Jiahan Li, Linghan Cai, Yongbing Zhang

IEEE Transactions on Medical Imaging · 2025-11 · 44 (11) · 4181-4197

DOI: 10.1109/tmi.2025.3569941 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Medical Imaging, 2025-11. Verfasst von Jiuyang Dong und 5 weiteren Autor:innen. Hauptthemen laut MEDLINE: Image Interpretation, Computer-Assisted, Machine Learning. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 49 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Shenzhen Science and Technology Project.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

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.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Image Interpretation, Computer-AssistedMachine LearningHumansAlgorithmsNeural Networks, ComputerMultiple-Instance Learning Algorithms

Förderung

  • National Natural Science Foundation of China
  • Shenzhen Science and Technology Project

Zitieren

APA

Dong, J., Jiang, J., Jiang, K., Li, J., Cai, L., & Zhang, Y. (2025). Disentangled Pseudo-Bag Augmentation for Whole Slide Image Multiple Instance Learning. IEEE Transactions on Medical Imaging, 44(11), 4181-4197. https://doi.org/10.1109/tmi.2025.3569941

Vancouver

Dong J, Jiang J, Jiang K, Li J, Cai L, Zhang Y. Disentangled Pseudo-Bag Augmentation for Whole Slide Image Multiple Instance Learning. IEEE Transactions on Medical Imaging. 2025;44(11):4181-4197. doi:10.1109/tmi.2025.3569941

BibTeX
@article{Dong2025,
  author = {Dong, Jiuyang and Jiang, Junjun and Jiang, Kui and Li, Jiahan and Cai, Linghan and Zhang, Yongbing},
  title = {Disentangled Pseudo-Bag Augmentation for Whole Slide Image Multiple Instance Learning},
  journal = {IEEE Transactions on Medical Imaging},
  year = {2025},
  volume = {44},
  number = {11},
  pages = {4181--4197},
  doi = {10.1109/tmi.2025.3569941}
}
Kostenlos starten →