Current Pathology Foundation Models are unrobust to Medical Center Differences

TL;DR

This study evaluates the robustness of current pathology foundation models to variations across medical centers, introducing a new metric called the Robustness Index to quantify the dominance of biological versus confounding features.

Contribution

The paper introduces the Robustness Index to measure model robustness to medical center differences and evaluates ten pathology models, revealing most are heavily influenced by confounding center signatures.

Findings

All models show strong influence of medical center signatures.

Only one model has a robustness index above one, indicating slight dominance of biological features.

Medical center origin is more accurately predicted than tissue source or cancer type.

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.