Pooling abnormal-enriched public mammography datasets into a screening cohort's training set lowers screening AUC, and dataset-of-origin is almost perfectly predictable despite identical preprocessing.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Dataset-Origin Signatures and Shortcut Learning in Screening Mammography AI: A Cross-Dataset Case Study
Pooling abnormal-enriched public mammography datasets into a screening cohort's training set lowers screening AUC, and dataset-of-origin is almost perfectly predictable despite identical preprocessing.