Foreground-only histogram matching
1 paper1 finding
- In silico1
Outcomes
1Breast density classification accuracy
upin silico1
1 study
Breast density classification accuracy
upin silico1
Foreground-only histogram matching improves breast density classification accuracy in silico (peak AUC of 98.32%).
“with the Swin Transformer backbone achieving a peak AUC of 98.32% for density classification.”
In silico
BreastMammo and DenseMammo: Benchmarks for Mammography Domain Generalization
arXiv10 Aug 2026Preprint
Latest
1 in 12 monthsarXiv
BreastMammo and DenseMammo: Benchmarks for Mammography Domain Generalization
Preprint: computational studyAI & data