Table 2.
Accuracy of automatically extracted imaging findings compared to manual gold standard review
| Imaging | Finding | Prevalence (%) | Precision (95 % CI) | Recall (95 % CI) | F measure |
|---|---|---|---|---|---|
| Screening mammography N = 200 | Positive BI-RADS | 26/79 (33%) | 1.0 (0.9, 1.0) | 1.0 (0.9, 1.0) | 1.0 |
| Calcification | 6/200 (3%) | 0.3 (0.1, 0.5) | 1.0 (0.5, 1.0) | 0.4 | |
| Mass | 4/200 (2%) | 0.6 (0.1, 0.9) | 0.8 (0.2, 1.0) | 0.7 | |
| Architectural distortiona | 7/200 (4%) | 1.0 (0.6, 1.0) | 1.0 (0.6, 1.0) | 1.0 | |
| Asymmetry | 15/200 (8%) | 0.8 (0.5, 1.0) | 0.7 (0.4, 0.9) | 0.7 | |
| Diagnostic mammography N = 182 | Positive BI-RADS | 13/67 (19%) | 1.0 (0.8, 1.0) | 0.9 (0.8, 1.0) | 0.9 |
| Calcification | 27/182 (15%) | 0.5 (0.3, 0.6) | 0.8 (0.6, 0.9) | 0.6 | |
| Mass | 24/182 (13%) | 0.6 (0.4, 0.8) | 0.8 (0.6, 1.0) | 0.7 | |
| Architectural distortiona | 7/182 (4%) | 0.3 (0.0, 0.9) | 0.1 (0.0, 0.6) | 0.2 | |
| Asymmetry | 21/182 (12%) | 0.4 (0.2, 0.7) | 0.5 (0.3, 0.7) | 0.5 | |
| MRI N = 196 | Positive BI-RADS | 123/193 (64%) | 0.7 (0.6, 0.8) | 0.7 (0.6, 0.8) | 0.7 |
| Mass | 99/196 (51%) | 0.8 (0.7, 0.9) | 0.9 (0.8, 0.9) | 0.8 | |
| Cysts | 16/196 (8%) | 0.6 (0.4, 0.8) | 1.0 (0.8, 1.0) | 0.8 | |
| NME | 36/196 (18%) | 0.9 (0.8, 1.0) | 0.9 (0.8, 1.0) | 0.9 | |
| Focus | 34/196 (17%) | 0.7 (0.5, 0.9) | 0.6 (0.4, 0.8) | 0.7 | |
| US N = 195 | Positive BI-RADS | 55/133 (41%) | 1.0 (0.9, 1.0) | 0.9 (0.8, 1.0) | 0.9 |
| Mass | 34/195 (17%) | 0.5 (0.4, 0.6) | 0.9 (0.8, 1.0) | 0.6 | |
| Cysts | 47/195 (24%) | 0.9 (0.7, 0.9) | 0.9 (0.8, 1.0) | 0.9 | |
| Architectural distortiona | 5/195 (3%) | 0.4 (0.1, 0.9) | 0.4 (0.1, 0.9) | 0.4 |
aExcluded from analysis