# Do Histopathological Foundation Models Eliminate Batch Effects? A Comparative Study

arXiv: https://arxiv.org/abs/2411.05489 · HTML: https://mlsanity.com/paper/2411.05489/

## Key facts

- Cited by 18 papers in the map (data through 14 Oct 2026).
- Builds on 4 of its 42 references that are in the map.
- First posted on arXiv 8 Nov 2024.

## Builds on (most cited first, 4 of 4)

- [Vision Models Are More Robust And Fair When Pretrained On Uncurated Images Without Supervision](https://mlsanity.com/paper/2202.08360/) (2022)
- [A Clinical Benchmark of Public Self-Supervised Pathology Foundation Models](https://mlsanity.com/paper/2407.06508/) (2024)
- [The Clever Hans Effect in Unsupervised Learning](https://mlsanity.com/paper/2408.08041/) (2024)
- [Towards Large-Scale Training of Pathology Foundation Models](https://mlsanity.com/paper/2404.15217/) (2024)

## Built on by (most cited first, 18 of 18)

- [Scanner-Induced Domain Shifts Undermine the Robustness of Pathology Foundation Models](https://mlsanity.com/paper/2601.04163/) (2026)
- [MeDi: Metadata-Guided Diffusion Models for Mitigating Biases in Tumor Classification](https://mlsanity.com/paper/2506.17140/) (2025)
- [Atlas 2 -- Foundation models for clinical deployment](https://mlsanity.com/paper/2601.05148/) (2026)
- [Beyond Attention Heatmaps: How to Get Better Explanations for Multiple Instance Learning Models in Histopathology](https://mlsanity.com/paper/2603.08328/) (2026)
- [MVHybrid: Improving Spatial Transcriptomics Prediction with Hybrid State Space-Vision Transformer Backbone in Pathology Vision Foundation Models](https://mlsanity.com/paper/2508.00383/) (2025)
- [AdvDINO: Domain-Adversarial Self-Supervised Representation Learning for Spatial Proteomics](https://mlsanity.com/paper/2508.04955/) (2025)
- [A protocol for evaluating robustness to H&E staining variation in computational pathology models](https://mlsanity.com/paper/2603.12886/) (2026)
- [Open-source framework for detecting bias and overfitting for large pathology images](https://mlsanity.com/paper/2503.01827/) (2025)
- [Comparing Computational Pathology Foundation Models using Representational Similarity Analysis](https://mlsanity.com/paper/2509.15482/) (2025)
- [Do Pathology Foundation Models Encode Disease Progression? A Pseudotime Analysis of Visual Representations](https://mlsanity.com/paper/2601.21334/) (2026)
- [Validation of an AI-based end-to-end model for prostate pathology using long-term archived routine samples](https://mlsanity.com/paper/2605.02614/) (2026)
- [When Are Multimodal Predictions Biologically Supported? A Diagnostic Evaluation Framework](https://mlsanity.com/paper/2605.31504/) (2026)
- [Mitigating Batch Effects in Histopathology via Language-Mediated Robust Embedding Generation](https://mlsanity.com/paper/2606.28697/) (2026)
- [A Distributional Robustness Margin For Pathology Foundation Models](https://mlsanity.com/paper/2607.25497/) (2026)
- [Harnessing Adversarial Distillation to Customise Debiased, Disease-Specific Pathology Foundation Models for Breast Cancer](https://mlsanity.com/paper/2608.01356/) (2026)
- [Multimodal LLMs Outperform Pathology Foundation Models in Cross-Domain Histological Similarity](https://mlsanity.com/paper/2609.32876/) (2026)
- [HERO: Histology Encoder for Robust Representation in Oncology](https://mlsanity.com/paper/2609.35943/) (2026)
- [Dataset-Specific Encoding of Image Content Underlies Cross-Dataset Failure in Brain MRI Foundation Models](https://mlsanity.com/paper/2610.13027/) (2026)

## Abstract (arXiv)

Deep learning has led to remarkable advancements in computational histopathology, e.g., in diagnostics, biomarker prediction, and outcome prognosis. Yet, the lack of annotated data and the impact of batch effects, e.g., systematic technical data differences across hospitals, hamper model robustness and generalization. Recent histopathological foundation models -- pretrained on millions to billions of images -- have been reported to improve generalization performances on various downstream tasks. However, it has not been systematically assessed whether they fully eliminate batch effects. In this study, we empirically show that the feature embeddings of the foundation models still contain distinct hospital signatures that can lead to biased predictions and misclassifications. We further find that the signatures are not removed by stain normalization methods, dominate distances in feature space, and are evident across various principal components. Our work provides a novel perspective on the evaluation of medical foundation models, paving the way for more robust pretraining strategies and downstream predictors.