StainNet: Scaling Self-Supervised Foundation Models on Immunohistochemistry and Special Stains for Computational Pathology

Jiawen Li, Jiali Hu, Xitong Ling, Yongqiang Lv, Yuxuan Chen, Yizhi Wang, Tian Guan, Yifei Liu, Yonghong He
Proceedings of The 9th International Conference on Medical Imaging with Deep Learning, PMLR 315:544-569, 2026.

Abstract

Foundation models trained with self-supervised learning (SSL) on large-scale histological images have significantly accelerated the development of computational pathology. These models can serve as backbones for region-of-interest (ROI) image analysis or patch-level feature extractors in whole-slide images (WSIs) based on multiple instance learning (MIL). Existing pathology foundation models (PFMs) are typically pre-trained on Hematoxylin-Eosin (H&E) stained pathology images. However, images such as immunohistochemistry (IHC) and special stains are also frequently used in clinical practice. PFMs pre-trained mainly on H&E-stained images may be limited in clinical applications involving these non-H&E images. To address this issue, we propose StainNet, a a collection of self-supervised foundation models specifically trained for IHC and special stains in pathology images based on the vision transformer (ViT) architecture. StainNet contains a ViT-Small and a ViT-Base model, both of which are trained using a self-distillation SSL approach on over 1.4 million patch images extracted from 20,231 publicly available IHC and special staining WSIs in the HISTAI database. To evaluate StainNet models, we conduct experiments on three in-house slide-level IHC classification tasks, three in-house ROI-level special stain and two public ROI-level IHC classification tasks to demonstrate their strong ability. We also perform ablation studies such as few-ratio learning and retrieval evaluations, and compare StainNet models with recent larger PFMs to further highlight their strengths.

Cite this Paper

BibTeX
@InProceedings{pmlr-v315-li26c, title = {StainNet: Scaling Self-Supervised Foundation Models on Immunohistochemistry and Special Stains for Computational Pathology}, author = {Li, Jiawen and Hu, Jiali and Ling, Xitong and Lv, Yongqiang and Chen, Yuxuan and Wang, Yizhi and Guan, Tian and Liu, Yifei and He, Yonghong}, booktitle = {Proceedings of The 9th International Conference on Medical Imaging with Deep Learning}, pages = {544--569}, year = {2026}, editor = {Huo, Yuankai and Gao, Mingchen and Kuo, Chang-Fu and Jin, Yueming and Deng, Ruining}, volume = {315}, series = {Proceedings of Machine Learning Research}, month = {08--10 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v315/main/assets/li26c/li26c.pdf}, url = {https://proceedings.mlr.press/v315/li26c.html}, abstract = {Foundation models trained with self-supervised learning (SSL) on large-scale histological images have significantly accelerated the development of computational pathology. These models can serve as backbones for region-of-interest (ROI) image analysis or patch-level feature extractors in whole-slide images (WSIs) based on multiple instance learning (MIL). Existing pathology foundation models (PFMs) are typically pre-trained on Hematoxylin-Eosin (H&E) stained pathology images. However, images such as immunohistochemistry (IHC) and special stains are also frequently used in clinical practice. PFMs pre-trained mainly on H&E-stained images may be limited in clinical applications involving these non-H&E images. To address this issue, we propose StainNet, a a collection of self-supervised foundation models specifically trained for IHC and special stains in pathology images based on the vision transformer (ViT) architecture. StainNet contains a ViT-Small and a ViT-Base model, both of which are trained using a self-distillation SSL approach on over 1.4 million patch images extracted from 20,231 publicly available IHC and special staining WSIs in the HISTAI database. To evaluate StainNet models, we conduct experiments on three in-house slide-level IHC classification tasks, three in-house ROI-level special stain and two public ROI-level IHC classification tasks to demonstrate their strong ability. We also perform ablation studies such as few-ratio learning and retrieval evaluations, and compare StainNet models with recent larger PFMs to further highlight their strengths.} }
Endnote
%0 Conference Paper %T StainNet: Scaling Self-Supervised Foundation Models on Immunohistochemistry and Special Stains for Computational Pathology %A Jiawen Li %A Jiali Hu %A Xitong Ling %A Yongqiang Lv %A Yuxuan Chen %A Yizhi Wang %A Tian Guan %A Yifei Liu %A Yonghong He %B Proceedings of The 9th International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2026 %E Yuankai Huo %E Mingchen Gao %E Chang-Fu Kuo %E Yueming Jin %E Ruining Deng %F pmlr-v315-li26c %I PMLR %P 544--569 %U https://proceedings.mlr.press/v315/li26c.html %V 315 %X Foundation models trained with self-supervised learning (SSL) on large-scale histological images have significantly accelerated the development of computational pathology. These models can serve as backbones for region-of-interest (ROI) image analysis or patch-level feature extractors in whole-slide images (WSIs) based on multiple instance learning (MIL). Existing pathology foundation models (PFMs) are typically pre-trained on Hematoxylin-Eosin (H&E) stained pathology images. However, images such as immunohistochemistry (IHC) and special stains are also frequently used in clinical practice. PFMs pre-trained mainly on H&E-stained images may be limited in clinical applications involving these non-H&E images. To address this issue, we propose StainNet, a a collection of self-supervised foundation models specifically trained for IHC and special stains in pathology images based on the vision transformer (ViT) architecture. StainNet contains a ViT-Small and a ViT-Base model, both of which are trained using a self-distillation SSL approach on over 1.4 million patch images extracted from 20,231 publicly available IHC and special staining WSIs in the HISTAI database. To evaluate StainNet models, we conduct experiments on three in-house slide-level IHC classification tasks, three in-house ROI-level special stain and two public ROI-level IHC classification tasks to demonstrate their strong ability. We also perform ablation studies such as few-ratio learning and retrieval evaluations, and compare StainNet models with recent larger PFMs to further highlight their strengths.
APA
Li, J., Hu, J., Ling, X., Lv, Y., Chen, Y., Wang, Y., Guan, T., Liu, Y. & He, Y.. (2026). StainNet: Scaling Self-Supervised Foundation Models on Immunohistochemistry and Special Stains for Computational Pathology. Proceedings of The 9th International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 315:544-569 Available from https://proceedings.mlr.press/v315/li26c.html.

Related Material