# Scanner-Induced Domain Shifts Undermine the Robustness of Pathology Foundation Models

arXiv: https://arxiv.org/abs/2601.04163 · HTML: https://mlsanity.com/paper/2601.04163/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Builds on 28 of its 53 references that are in the map.
- First posted on arXiv 7 Jan 2026.

## Builds on (most cited first, 28 of 28)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction](https://mlsanity.com/paper/1802.03426/) (2018)
- [CoCa: Contrastive Captioners are Image-Text Foundation Models](https://mlsanity.com/paper/2205.01917/) (2022)
- [Attention-based Deep Multiple Instance Learning](https://mlsanity.com/paper/1802.04712/) (2018)
- [iBOT: Image BERT Pre-Training with Online Tokenizer](https://mlsanity.com/paper/2111.07832/) (2021)
- [Data Efficient and Weakly Supervised Computational Pathology on Whole Slide Images](https://mlsanity.com/paper/2004.09666/) (2020)
- [Quantifying the effects of data augmentation and stain color normalization in convolutional neural networks for computational pathology](https://mlsanity.com/paper/1902.06543/) (2019)
- [Deep neural network models for computational histopathology: A survey](https://mlsanity.com/paper/1912.12378/) (2019)
- [Multimodal Whole Slide Foundation Model for Pathology](https://mlsanity.com/paper/2411.19666/) (2024)
- [Virchow2: Scaling Self-Supervised Mixed Magnification Models in Pathology](https://mlsanity.com/paper/2408.00738/) (2024)
- [Artificial Intelligence for Digital and Computational Pathology](https://mlsanity.com/paper/2401.06148/) (2023)
- [Phikon-v2, A large and public feature extractor for biomarker prediction](https://mlsanity.com/paper/2409.09173/) (2024)
- [A Clinical Benchmark of Public Self-Supervised Pathology Foundation Models](https://mlsanity.com/paper/2407.06508/) (2024)
- [Benchmarking foundation models as feature extractors for weakly-supervised computational pathology](https://mlsanity.com/paper/2408.15823/) (2024)
- [Domain Generalization in Computational Pathology: Survey and Guidelines](https://mlsanity.com/paper/2310.19656/) (2023)
- [Foundation Models in Computational Pathology: A Review of Challenges, Opportunities, and Impact](https://mlsanity.com/paper/2502.08333/) (2025)
- [The importance of stain normalization in colorectal tissue classification with convolutional networks](https://mlsanity.com/paper/1702.05931/) (2017)
- [Do Histopathological Foundation Models Eliminate Batch Effects? A Comparative Study](https://mlsanity.com/paper/2411.05489/) (2024)
- [A Survey on Computational Pathology Foundation Models: Datasets, Adaptation Strategies, and Evaluation Tasks](https://mlsanity.com/paper/2501.15724/) (2025)
- [A Comprehensive Evaluation of Histopathology Foundation Models for Ovarian Cancer Subtype Classification](https://mlsanity.com/paper/2405.09990/) (2024)
- [Pathology Foundation Models are Scanner Sensitive: Benchmark and Mitigation with Contrastive ScanGen Loss](https://mlsanity.com/paper/2507.22092/) (2025)
- [Hospital-Agnostic Image Representation Learning in Digital Pathology](https://mlsanity.com/paper/2204.02404/) (2022)
- [Are the Latent Representations of Foundation Models for Pathology Invariant to Rotation?](https://mlsanity.com/paper/2412.11938/) (2024)
- [SCORPION: Addressing Scanner-Induced Variability in Histopathology](https://mlsanity.com/paper/2507.20907/) (2025)

## Built on by (most cited first, 9 of 9)

- [CytoSyn: a Foundation Diffusion Model for Histopathology -- Tech Report](https://mlsanity.com/paper/2603.18089/) (2026)
- [Benchmarking Pathology Foundation Models for Breast Cancer Survival Prediction](https://mlsanity.com/paper/2604.24679/) (2026)
- [Autonomous Drift Learning in Data Streams: A Unified Perspective](https://mlsanity.com/paper/2605.01295/) (2026)
- [Discrepancy Minimization Improves Cross-Hospital Robustness in Digital Pathology](https://mlsanity.com/paper/2605.25175/) (2026)
- [When Are Multimodal Predictions Biologically Supported? A Diagnostic Evaluation Framework](https://mlsanity.com/paper/2605.31504/) (2026)
- [Robustifying pathology foundation models via fine-tuning](https://mlsanity.com/paper/2607.22861/) (2026)
- [An end-to-end-trained vision-language model for native-language prostate pathology report generation](https://mlsanity.com/paper/2608.23143/) (2026)
- [Can You Trust Frozen Hematology Foundation Models under Acquisition Shift?](https://mlsanity.com/paper/2608.25148/) (2026)
- [Multimodal LLMs Outperform Pathology Foundation Models in Cross-Domain Histological Similarity](https://mlsanity.com/paper/2609.32876/) (2026)

## Abstract (arXiv)

Pathology foundation models (PFMs) have become central to computational pathology, aiming to offer general encoders for feature extraction from whole-slide images (WSIs). Despite strong benchmark performance, PFM robustness to real-world technical domain shifts, such as variability from whole-slide scanner devices, remains poorly understood. We systematically evaluated the robustness of 14 PFMs to scanner-induced variability, including state-of-the-art models, earlier self-supervised models, and a baseline trained on natural images. Using a multiscanner dataset of 384 breast cancer WSIs scanned on five devices, we isolated scanner effects independently from biological and laboratory confounders. Robustness is assessed via complementary unsupervised embedding analyses and a set of clinicopathological supervised prediction tasks. Our results demonstrate that current PFMs are not invariant to scanner-induced domain shifts. Most models encode pronounced scanner-specific variability in their embedding spaces. While AUC often remains stable, this masks a critical failure mode: scanner variability systematically alters the embedding space and impacts calibration of downstream model predictions, resulting in scanner-dependent bias that can impact reliability in clinical use cases. We further show that robustness is not a simple function of training data scale, model size, or model recency. None of the models provided reliable robustness against scanner-induced variability. While the models trained on the most diverse data, here represented by vision-language models, appear to have an advantage with respect to robustness, they underperformed on downstream supervised tasks. We conclude that development and evaluation of PFMs requires moving beyond accuracy-centric benchmarks toward explicit evaluation and optimisation of embedding stability and calibration under realistic acquisition variability.