Abstract:Stain variation across hospitals degrades histopathology models at deployment. Existing augmentation methods perturb color spaces with arbitrary hyperparameters, lacking both a principled budget and coverage guarantees for unseen centers. We propose \textbf{C}alibrated \textbf{A}dversarial \textbf{S}tain \textbf{A}ugmentation (\textbf{CASA}), which performs adversarial augmentation in the Macenko stain parameter space with a budget calibrated from multi-center statistics via the DKW inequality. On Camelyon17-WILDS (5 seeds), CASA achieves $93.9\% \pm 1.6\%$ slide-level accuracy -- outperforming HED-strong ($88.4\% \pm 7.3\%$), RandStainNA ($85.2\% \pm 6.7\%$), and ERM ($63.9\% \pm 11.3\%$) -- with the highest worst-group accuracy ($84.9\% \pm 0.9\%$) among all 10 compared methods.
| Subjects: | Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG) |
| Cite as: | arXiv:2605.13889 [eess.IV] |
| (or arXiv:2605.13889v1 [eess.IV] for this version) | |
| https://doi.org/10.48550/arXiv.2605.13889 arXiv-issued DOI via DataCite |
Submission history
From: Mingi Hong [view email]
[v1]
Tue, 12 May 2026 04:39:33 UTC (42 KB)