Learning Disentangled Stain and Structural Representations for Semi-Supervised Histopathology Segmentation

Ha-Hieu Pham, Nguyen Lan Vi Vu, Thanh-Huy Nguyen, Ulas Bagci, Min Xu, Trung-Nghia Le, Huy-Hieu Pham
Proceedings of the MICCAI Workshop on Computational Pathology, PMLR 316:278-287, 2026.

Abstract

Accurate gland segmentation in histopathology images is essential for cancer diagnosis and prognosis. However, significant variability in Hematoxylin and Eosin (H&E) staining and tissue morphology, combined with limited annotated data, poses major challenges for automated segmentation. To address this, we propose Color-Structure Dual-Student (CSDS), a novel semi-supervised segmentation framework designed to learn disentangled representations of stain appearance and tissue structure. CSDS comprises two specialized student networks: one trained on stain-augmented inputs to model chromatic variation, and the other on structure-augmented inputs to capture morphological cues. A shared teacher network, updated via Exponential Moving Average (EMA), supervises both students through pseudo-labels. To further improve label reliability, we introduce stain-aware and structureaware uncertainty estimation modules that adaptively modulate the contribution of each student during training. Experiments on the GlaS and CRAG datasets show that CSDS achieves state-of-the-art performance in low-label settings, with Dice score improvements of up to 1.2% on GlaS and 0.7% on CRAG at 5% labeled data, and 0.7% and 1.4% at 10%. Our code and pre-trained models are available at https://github.com/hieuphamha19/CSDS.

Cite this Paper

BibTeX
@InProceedings{pmlr-v316-pham26a, title = {Learning Disentangled Stain and Structural Representations for Semi-Supervised Histopathology Segmentation}, author = {Pham, Ha-Hieu and Vu, Nguyen Lan Vi and Nguyen, Thanh-Huy and Bagci, Ulas and Xu, Min and Le, Trung-Nghia and Pham, Huy-Hieu}, booktitle = {Proceedings of the MICCAI Workshop on Computational Pathology}, pages = {278--287}, 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/pham26a/pham26a.pdf}, url = {https://proceedings.mlr.press/v316/pham26a.html}, abstract = {Accurate gland segmentation in histopathology images is essential for cancer diagnosis and prognosis. However, significant variability in Hematoxylin and Eosin (H&E) staining and tissue morphology, combined with limited annotated data, poses major challenges for automated segmentation. To address this, we propose Color-Structure Dual-Student (CSDS), a novel semi-supervised segmentation framework designed to learn disentangled representations of stain appearance and tissue structure. CSDS comprises two specialized student networks: one trained on stain-augmented inputs to model chromatic variation, and the other on structure-augmented inputs to capture morphological cues. A shared teacher network, updated via Exponential Moving Average (EMA), supervises both students through pseudo-labels. To further improve label reliability, we introduce stain-aware and structureaware uncertainty estimation modules that adaptively modulate the contribution of each student during training. Experiments on the GlaS and CRAG datasets show that CSDS achieves state-of-the-art performance in low-label settings, with Dice score improvements of up to 1.2% on GlaS and 0.7% on CRAG at 5% labeled data, and 0.7% and 1.4% at 10%. Our code and pre-trained models are available at https://github.com/hieuphamha19/CSDS.} }
Endnote
%0 Conference Paper %T Learning Disentangled Stain and Structural Representations for Semi-Supervised Histopathology Segmentation %A Ha-Hieu Pham %A Nguyen Lan Vi Vu %A Thanh-Huy Nguyen %A Ulas Bagci %A Min Xu %A Trung-Nghia Le %A Huy-Hieu Pham %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-pham26a %I PMLR %P 278--287 %U https://proceedings.mlr.press/v316/pham26a.html %V 316 %X Accurate gland segmentation in histopathology images is essential for cancer diagnosis and prognosis. However, significant variability in Hematoxylin and Eosin (H&E) staining and tissue morphology, combined with limited annotated data, poses major challenges for automated segmentation. To address this, we propose Color-Structure Dual-Student (CSDS), a novel semi-supervised segmentation framework designed to learn disentangled representations of stain appearance and tissue structure. CSDS comprises two specialized student networks: one trained on stain-augmented inputs to model chromatic variation, and the other on structure-augmented inputs to capture morphological cues. A shared teacher network, updated via Exponential Moving Average (EMA), supervises both students through pseudo-labels. To further improve label reliability, we introduce stain-aware and structureaware uncertainty estimation modules that adaptively modulate the contribution of each student during training. Experiments on the GlaS and CRAG datasets show that CSDS achieves state-of-the-art performance in low-label settings, with Dice score improvements of up to 1.2% on GlaS and 0.7% on CRAG at 5% labeled data, and 0.7% and 1.4% at 10%. Our code and pre-trained models are available at https://github.com/hieuphamha19/CSDS.
APA
Pham, H., Vu, N.L.V., Nguyen, T., Bagci, U., Xu, M., Le, T. & Pham, H.. (2026). Learning Disentangled Stain and Structural Representations for Semi-Supervised Histopathology Segmentation. Proceedings of the MICCAI Workshop on Computational Pathology, in Proceedings of Machine Learning Research 316:278-287 Available from https://proceedings.mlr.press/v316/pham26a.html.

Related Material