Transformer-based multiple instance learning network with 2D positional encoding for histopathology image classification

dc.citation.issue5
dc.citation.volume11
dc.contributor.authorYang B
dc.contributor.authorDing L
dc.contributor.authorLi J
dc.contributor.authorLi Y
dc.contributor.authorQu G
dc.contributor.authorWang J
dc.contributor.authorWang Q
dc.contributor.authorLiu B
dc.date.accessioned2025-05-27T23:48:59Z
dc.date.available2025-05-27T23:48:59Z
dc.date.issued2025-05
dc.description.abstractDigital medical imaging, particularly pathology images, is essential for cancer diagnosis but faces challenges in direct model training due to its super-resolution nature. Although weakly supervised learning has reduced the need for manual annotations, many multiple instance learning (MIL) methods struggle to effectively capture crucial spatial relationships in histopathological images. Existing methods incorporating positional information often overlook nuanced spatial correlations or use positional encoding strategies that do not fully capture the unique spatial dynamics of pathology images. To address this issue, we propose a new framework named TMIL (Transformer-based Multiple Instance Learning Network with 2D positional encoding), which leverages multiple instance learning for weakly supervised classification of histopathological images. TMIL incorporates a 2D positional encoding module, based on the Transformer, to model positional information and explore correlations between instances. Furthermore, TMIL divides histopathological images into pseudo-bags and trains patch-level feature vectors with deep metric learning to enhance classification performance. Finally, the proposed approach is evaluated on a public colorectal adenoma dataset. The experimental results show that TMIL outperforms existing MIL methods, achieving an AUC of 97.28% and an ACC of 95.19%. These findings suggest that TMIL’s integration of deep metric learning and positional encoding offers a promising approach for improving the efficiency and accuracy of pathology image analysis in cancer diagnosis.
dc.description.confidentialfalse
dc.edition.editionMay 2025
dc.identifier.citationYang B, Ding L, Li J, Li Y, Qu G, Wang J, Wang Q, Liu B. (2025). Transformer-based multiple instance learning network with 2D positional encoding for histopathology image classification. Complex and Intelligent Systems. 11. 5.
dc.identifier.doi10.1007/s40747-025-01779-y
dc.identifier.eissn2198-6053
dc.identifier.elements-typejournal-article
dc.identifier.issn2199-4536
dc.identifier.number218
dc.identifier.piis40747-025-01779-y
dc.identifier.urihttps://mro.massey.ac.nz/handle/10179/72953
dc.languageEnglish
dc.publisherSpringer Nature Switzerland AG
dc.publisher.urihttps://link.springer.com/article/10.1007/s40747-025-01779-y
dc.relation.isPartOfComplex and Intelligent Systems
dc.rights(c) 2025 The Author/s
dc.rightsCC BY-NC-ND 4.0
dc.rights.urihttps://creativecommons.org/licenses/by-nc-nd/4.0/
dc.subjectWeakly supervised training
dc.subjectImage classification
dc.subjectMultiple instance learning
dc.titleTransformer-based multiple instance learning network with 2D positional encoding for histopathology image classification
dc.typeJournal article
pubs.elements-id500303
pubs.organisational-groupOther
Files
Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
500303 PDF.pdf
Size:
2.92 MB
Format:
Adobe Portable Document Format
Description:
Preprint version.pdf
License bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
license.txt
Size:
9.22 KB
Format:
Plain Text
Description:
Collections