Table 3 Subjects and concepts included in the homogenized dataset
From: Public data homogenization for AI model development in breast cancer
Dataset | Subjects included | Mapped concepts | ||
|---|---|---|---|---|
Number | Percentage | Number | Percentage | |
ISPY2 | 719 | 73% | 12 | 133% |
Duke-Breast-Cancer-MRI | 922 | 100% | 36 | 49% |
ISPY1 | 221 | 100% | 14 | 82% |
TCGA-BRCA | 139 | 99% | 27 | 25% |
Breast-MRI-NACT-Pilot | 64 | 100% | 23 | 56% |