Context-guided Prompt Learning for Continual WSI Classification

Giulia Corso, Francesca Miccolis, Angelo Porrello, Federico Bolelli, Simone Calderara, Elisa Ficarra
Proceedings of the MICCAI Workshop on Computational Pathology, PMLR 316:187-198, 2026.

Abstract

Whole Slide Images (WSIs) are crucial in histological diagnostics, providing high-resolution insights into cellular structures. In addition to challenges like the gigapixel scale of WSIs and the lack of pixel-level annotations, privacy restrictions further complicate their analysis. For instance, in a hospital network, different facilities need to collaborate on WSI analysis without the possibility of sharing sensitive patient data. A more practical and secure approach involves sharing models capable of continual adaptation to new data. However, without proper measures, catastrophic forgetting can occur. Traditional continual learning techniques rely on storing previous data, which violates privacy restrictions. To address this issue, this paper introduces Context Optimization Multiple Instance Learning (CooMIL), a rehearsal-free continual learning framework explicitly designed for WSI analysis. It employs a WSI-specific prompt learning procedure to adapt classification models across tasks, efficiently preventing catastrophic forgetting. Evaluated on four public WSI datasets from TCGA projects, our model significantly outperforms state-of-the-art methods within the WSI-based continual learning framework. The source code is available at https://github.com/FrancescaMiccolis/CooMIL.

Cite this Paper

BibTeX
@InProceedings{pmlr-v316-corso26a, title = {Context-guided Prompt Learning for Continual WSI Classification}, author = {Corso, Giulia and Miccolis, Francesca and Porrello, Angelo and Bolelli, Federico and Calderara, Simone and Ficarra, Elisa}, booktitle = {Proceedings of the MICCAI Workshop on Computational Pathology}, pages = {187--198}, year = {2026}, editor = {Studer, Linda and Ciompi, Francesco and Khalili, Nadieh and Faryna, Khrystyna and Faryna, Khrystyna and Yeong, Joe and Lau, Mai Chan and Chen, Hao and Liu, Ziyi and Brattoli, Biagio}, volume = {316}, series = {Proceedings of Machine Learning Research}, month = {27 Sep}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v316/main/assets/corso26a/corso26a.pdf}, url = {https://proceedings.mlr.press/v316/corso26a.html}, abstract = {Whole Slide Images (WSIs) are crucial in histological diagnostics, providing high-resolution insights into cellular structures. In addition to challenges like the gigapixel scale of WSIs and the lack of pixel-level annotations, privacy restrictions further complicate their analysis. For instance, in a hospital network, different facilities need to collaborate on WSI analysis without the possibility of sharing sensitive patient data. A more practical and secure approach involves sharing models capable of continual adaptation to new data. However, without proper measures, catastrophic forgetting can occur. Traditional continual learning techniques rely on storing previous data, which violates privacy restrictions. To address this issue, this paper introduces Context Optimization Multiple Instance Learning (CooMIL), a rehearsal-free continual learning framework explicitly designed for WSI analysis. It employs a WSI-specific prompt learning procedure to adapt classification models across tasks, efficiently preventing catastrophic forgetting. Evaluated on four public WSI datasets from TCGA projects, our model significantly outperforms state-of-the-art methods within the WSI-based continual learning framework. The source code is available at https://github.com/FrancescaMiccolis/CooMIL.} }
Endnote
%0 Conference Paper %T Context-guided Prompt Learning for Continual WSI Classification %A Giulia Corso %A Francesca Miccolis %A Angelo Porrello %A Federico Bolelli %A Simone Calderara %A Elisa Ficarra %B Proceedings of the MICCAI Workshop on Computational Pathology %C Proceedings of Machine Learning Research %D 2026 %E Linda Studer %E Francesco Ciompi %E Nadieh Khalili %E Khrystyna Faryna %E Khrystyna Faryna %E Joe Yeong %E Mai Chan Lau %E Hao Chen %E Ziyi Liu %E Biagio Brattoli %F pmlr-v316-corso26a %I PMLR %P 187--198 %U https://proceedings.mlr.press/v316/corso26a.html %V 316 %X Whole Slide Images (WSIs) are crucial in histological diagnostics, providing high-resolution insights into cellular structures. In addition to challenges like the gigapixel scale of WSIs and the lack of pixel-level annotations, privacy restrictions further complicate their analysis. For instance, in a hospital network, different facilities need to collaborate on WSI analysis without the possibility of sharing sensitive patient data. A more practical and secure approach involves sharing models capable of continual adaptation to new data. However, without proper measures, catastrophic forgetting can occur. Traditional continual learning techniques rely on storing previous data, which violates privacy restrictions. To address this issue, this paper introduces Context Optimization Multiple Instance Learning (CooMIL), a rehearsal-free continual learning framework explicitly designed for WSI analysis. It employs a WSI-specific prompt learning procedure to adapt classification models across tasks, efficiently preventing catastrophic forgetting. Evaluated on four public WSI datasets from TCGA projects, our model significantly outperforms state-of-the-art methods within the WSI-based continual learning framework. The source code is available at https://github.com/FrancescaMiccolis/CooMIL.
APA
Corso, G., Miccolis, F., Porrello, A., Bolelli, F., Calderara, S. & Ficarra, E.. (2026). Context-guided Prompt Learning for Continual WSI Classification. Proceedings of the MICCAI Workshop on Computational Pathology, in Proceedings of Machine Learning Research 316:187-198 Available from https://proceedings.mlr.press/v316/corso26a.html.

Related Material