Quantifying and Mitigating Hospital Domain Bias in Pathology Foundation Models using Adversarial Feature Disentanglement

Mengliang Zhang
Proceedings of The 9th International Conference on Medical Imaging with Deep Learning, PMLR 315:3866-3884, 2026.

Abstract

Pathology foundation models (PFMs) have demonstrated remarkable potential in whole-slide image (WSI) diagnosis. However, pathology images from different hospitals exhibit domain shifts due to variations in scanning hardware and preprocessing. These differences cause PFMs to learn spurious hospital-specific features, severely compromising their robustness and generalizability in clinical settings. We present the first systematic study of this hospital-source domain bias in PFMs. To address the critical trade-off between diagnostic utility and domain predictability, we establish a quantification pipeline and introduce the Robustness Index (RI). Furthermore, we propose a lightweight adversarial framework for feature disentanglement. This framework employs a trainable adapter and a domain classifier connected via a Gradient Reversal Layer (GRL) to remove latent hospital-specific information from frozen PFM representations without modifying the encoder itself. Experiments on multi-center histopathology datasets demonstrate that our approach substantially suppresses domain predictability and achieves significant gains in feature robustness. Crucially, the method maintains or improves disease classification performance, proving its efficacy particularly in out-of-domain scenarios.

Cite this Paper

BibTeX
@InProceedings{pmlr-v315-zhang26d, title = {Quantifying and Mitigating Hospital Domain Bias in Pathology Foundation Models using Adversarial Feature Disentanglement}, author = {Zhang, Mengliang}, booktitle = {Proceedings of The 9th International Conference on Medical Imaging with Deep Learning}, pages = {3866--3884}, 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/zhang26d/zhang26d.pdf}, url = {https://proceedings.mlr.press/v315/zhang26d.html}, abstract = {Pathology foundation models (PFMs) have demonstrated remarkable potential in whole-slide image (WSI) diagnosis. However, pathology images from different hospitals exhibit domain shifts due to variations in scanning hardware and preprocessing. These differences cause PFMs to learn spurious hospital-specific features, severely compromising their robustness and generalizability in clinical settings. We present the first systematic study of this hospital-source domain bias in PFMs. To address the critical trade-off between diagnostic utility and domain predictability, we establish a quantification pipeline and introduce the Robustness Index (RI). Furthermore, we propose a lightweight adversarial framework for feature disentanglement. This framework employs a trainable adapter and a domain classifier connected via a Gradient Reversal Layer (GRL) to remove latent hospital-specific information from frozen PFM representations without modifying the encoder itself. Experiments on multi-center histopathology datasets demonstrate that our approach substantially suppresses domain predictability and achieves significant gains in feature robustness. Crucially, the method maintains or improves disease classification performance, proving its efficacy particularly in out-of-domain scenarios.} }
Endnote
%0 Conference Paper %T Quantifying and Mitigating Hospital Domain Bias in Pathology Foundation Models using Adversarial Feature Disentanglement %A Mengliang Zhang %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-zhang26d %I PMLR %P 3866--3884 %U https://proceedings.mlr.press/v315/zhang26d.html %V 315 %X Pathology foundation models (PFMs) have demonstrated remarkable potential in whole-slide image (WSI) diagnosis. However, pathology images from different hospitals exhibit domain shifts due to variations in scanning hardware and preprocessing. These differences cause PFMs to learn spurious hospital-specific features, severely compromising their robustness and generalizability in clinical settings. We present the first systematic study of this hospital-source domain bias in PFMs. To address the critical trade-off between diagnostic utility and domain predictability, we establish a quantification pipeline and introduce the Robustness Index (RI). Furthermore, we propose a lightweight adversarial framework for feature disentanglement. This framework employs a trainable adapter and a domain classifier connected via a Gradient Reversal Layer (GRL) to remove latent hospital-specific information from frozen PFM representations without modifying the encoder itself. Experiments on multi-center histopathology datasets demonstrate that our approach substantially suppresses domain predictability and achieves significant gains in feature robustness. Crucially, the method maintains or improves disease classification performance, proving its efficacy particularly in out-of-domain scenarios.
APA
Zhang, M.. (2026). Quantifying and Mitigating Hospital Domain Bias in Pathology Foundation Models using Adversarial Feature Disentanglement. Proceedings of The 9th International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 315:3866-3884 Available from https://proceedings.mlr.press/v315/zhang26d.html.

Related Material