Attention-based Generative Latent Replay: A Continual Learning Approach for WSI Analysis

Pratibha Kumari, Daniel Reisenbüchler, Afshin Bozorgpour, Nadine S. Schaadt, Friedrich Feuerhake, Dorit Merhof
Proceedings of the MICCAI Workshop on Computational Pathology, PMLR 316:12-23, 2026.

Abstract

Whole slide image (WSI) classification has emerged as a powerful tool in computational pathology, but remains constrained by domain shifts, e.g., due to different organs, diseases, or institution-specific variations. To address this challenge, we propose an Attention-based Generative Latent Replay Continual Learning framework (AGLR-CL), in a multiple instance learning (MIL) setup for domain incremental WSI classification. Our method employs Gaussian Mixture Models (GMMs) to synthesize WSI representations and patch count distributions, preserving knowledge of past domains without explicitly storing original data. A novel attention-based filtering step focuses on the most salient patch embeddings, ensuring high-quality synthetic samples. This privacy-aware strategy obviates the need for replay buffers and outperforms other buffer-free counterparts while matching the performance of buffer-based solutions. We validate AGLR-CL on clinically relevant biomarker detection and molecular status prediction across multiple public datasets with diverse centers, organs, and patient cohorts. Experimental results confirm its ability to retain prior knowledge and adapt to new domains, offering an effective, privacy-preserving avenue for domain incremental continual learning in WSI classification.

Cite this Paper

BibTeX
@InProceedings{pmlr-v316-kumari26a, title = {Attention-based Generative Latent Replay: A Continual Learning Approach for WSI Analysis}, author = {Kumari, Pratibha and Reisenb\"{u}chler, Daniel and Bozorgpour, Afshin and Schaadt, Nadine S. and Feuerhake, Friedrich and Merhof, Dorit}, booktitle = {Proceedings of the MICCAI Workshop on Computational Pathology}, pages = {12--23}, 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/kumari26a/kumari26a.pdf}, url = {https://proceedings.mlr.press/v316/kumari26a.html}, abstract = {Whole slide image (WSI) classification has emerged as a powerful tool in computational pathology, but remains constrained by domain shifts, e.g., due to different organs, diseases, or institution-specific variations. To address this challenge, we propose an Attention-based Generative Latent Replay Continual Learning framework (AGLR-CL), in a multiple instance learning (MIL) setup for domain incremental WSI classification. Our method employs Gaussian Mixture Models (GMMs) to synthesize WSI representations and patch count distributions, preserving knowledge of past domains without explicitly storing original data. A novel attention-based filtering step focuses on the most salient patch embeddings, ensuring high-quality synthetic samples. This privacy-aware strategy obviates the need for replay buffers and outperforms other buffer-free counterparts while matching the performance of buffer-based solutions. We validate AGLR-CL on clinically relevant biomarker detection and molecular status prediction across multiple public datasets with diverse centers, organs, and patient cohorts. Experimental results confirm its ability to retain prior knowledge and adapt to new domains, offering an effective, privacy-preserving avenue for domain incremental continual learning in WSI classification.} }
Endnote
%0 Conference Paper %T Attention-based Generative Latent Replay: A Continual Learning Approach for WSI Analysis %A Pratibha Kumari %A Daniel Reisenbüchler %A Afshin Bozorgpour %A Nadine S. Schaadt %A Friedrich Feuerhake %A Dorit Merhof %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-kumari26a %I PMLR %P 12--23 %U https://proceedings.mlr.press/v316/kumari26a.html %V 316 %X Whole slide image (WSI) classification has emerged as a powerful tool in computational pathology, but remains constrained by domain shifts, e.g., due to different organs, diseases, or institution-specific variations. To address this challenge, we propose an Attention-based Generative Latent Replay Continual Learning framework (AGLR-CL), in a multiple instance learning (MIL) setup for domain incremental WSI classification. Our method employs Gaussian Mixture Models (GMMs) to synthesize WSI representations and patch count distributions, preserving knowledge of past domains without explicitly storing original data. A novel attention-based filtering step focuses on the most salient patch embeddings, ensuring high-quality synthetic samples. This privacy-aware strategy obviates the need for replay buffers and outperforms other buffer-free counterparts while matching the performance of buffer-based solutions. We validate AGLR-CL on clinically relevant biomarker detection and molecular status prediction across multiple public datasets with diverse centers, organs, and patient cohorts. Experimental results confirm its ability to retain prior knowledge and adapt to new domains, offering an effective, privacy-preserving avenue for domain incremental continual learning in WSI classification.
APA
Kumari, P., Reisenbüchler, D., Bozorgpour, A., Schaadt, N.S., Feuerhake, F. & Merhof, D.. (2026). Attention-based Generative Latent Replay: A Continual Learning Approach for WSI Analysis. Proceedings of the MICCAI Workshop on Computational Pathology, in Proceedings of Machine Learning Research 316:12-23 Available from https://proceedings.mlr.press/v316/kumari26a.html.

Related Material