Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Discrepancy Minimization Improves Cross-Hospital Robustness in Digital Pathology

About

Pathology foundation models (PFMs) have advanced rapidly in recent years and support training classifiers for a range of histopathology tasks. However, their robustness across hospitals remains limited: performance often degrades when training a classifier on data from one hospital and evaluating it on another target hospital. We address this challenge by fine-tuning PFMs with a local maximum mean discrepancy (LMMD) objective that applies to two settings: domain adaptation, where unlabeled target-hospital data is available, and domain generalization, where target-hospital data is unavailable at all. Experiments at both the patch- and slide-level show consistent improvements across multiple PFMs and tasks.

Ben Vardi, Dana Schonberger, Yuval Friedmann, Zohar Yakhini, Iris Barshack, Alexander Loebel, Ariel Shamir• 2026

Related benchmarks

TaskDatasetResultRank
Slide-level classificationTCGA NSCLC (test)--
96
Domain AdaptationNSCLC patch-level
Balanced Accuracy84.88
12
Domain AdaptationRCC patch-level
Balanced Accuracy96.85
10
Patch-Level ClassificationNSCLC Unseen target hospitals
Balanced Accuracy83.79
10
Slide-level classificationRCC Slide-level (test)
Balanced Accuracy97.17
10
Domain AdaptationRCC patch-level (Gen)
Balanced Accuracy94.1
2
Domain AdaptationNSCLC patch-level (Unbalanced)
Balanced Accuracy80.48
2
Domain AdaptationRCC patch-level (Unbalanced)
Balanced Accuracy94.4
2
Showing 8 of 8 rows

Other info

Follow for update