REVIEW 7 cited by
Current Pathology Foundation Models are unrobust to Medical Center Differences
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Pathology Foundation Models (FMs) hold great promise for healthcare. Before they can be used in clinical practice, it is essential to ensure they are robust to variations between medical centers. We measure whether pathology FMs focus on biological features like tissue and cancer type, or on the well known confounding medical center signatures introduced by staining procedure and other differences. We introduce the Robustness Index. This novel robustness metric reflects to what degree biological features dominate confounding features. Ten current publicly available pathology FMs are evaluated. We find that all current pathology foundation models evaluated represent the medical center to a strong degree. Significant differences in the robustness index are observed. Only one model so far has a robustness index greater than one, meaning biological features dominate confounding features, but only slightly. A quantitative approach to measure the influence of medical center differences on FM-based prediction performance is described. We analyze the impact of unrobustness on classification performance of downstream models, and find that cancer-type classification errors are not random, but specifically attributable to same-center confounders: images of other classes from the same medical center. We visualize FM embedding spaces, and find these are more strongly organized by medical centers than by biological factors. As a consequence, the medical center of origin is predicted more accurately than the tissue source and cancer type. The robustness index introduced here is provided with the aim of advancing progress towards clinical adoption of robust and reliable pathology FMs.
Forward citations
Cited by 7 Pith papers
-
Beyond Counts: A Distributional Robustness Margin For Pathology Foundation Models
CRoMa scores each pathology image embedding by the margin between cross-site biological matches and same-site biological distractors, revealing distributional lower tails that pooled robustness scores hide.
-
The Good, the Bad, and the Brittle: Benchmarking Robustness and Generalisation of Histopathology Foundation Models
Mid-sized pathology foundation models match or beat billion-parameter ones on clinically realistic perturbations and distribution-shift tests, so scaling alone has largely saturated for robustness.
-
Towards Robust Foundation Models for Digital Pathology
PathoROB shows that all 20 evaluated pathology foundation models encode medical center information and that lower robustness correlates with larger downstream performance drops.
-
HASD: Hierarchical Adaption for pathology Slide-level Domain-shift
A hierarchical domain adaptation framework improves slide-level HER2 grading and survival prediction across medical centers by aligning features at domain, slide, and patch levels.
-
MeDi: Metadata-Guided Diffusion Models for Mitigating Biases in Tumor Classification
Conditioning a histopathology diffusion model on tissue source site metadata improves synthetic image fidelity and enhances downstream tumor classification under subpopulation shift.
-
Atlas 2 -- Foundation models for clinical deployment
Atlas 2 and its distilled variants set new average state-of-the-art results across 80 pathology benchmarks, with larger robustness margins over prior models.
-
Pathology Foundation Models are Scanner Sensitive: Benchmark and Mitigation with Contrastive ScanGen Loss
Pathology foundation models produce scanner-dependent predictions, and the ScanGen contrastive loss reduces this scanner bias during fine-tuning for EGFR mutation prediction from whole slide images.
Discussion (0). Continue with ORCID to comment.