Table 5. The performance of the LR classifier in distinguishing between luminal and nonluminal breast cancer.
| Regions | Cohort | AUC (95% CI) | SEN (%) | SPE (%) | ACC (%) | PPV (%) | NPV (%) |
|---|---|---|---|---|---|---|---|
| Rapid | Training | 0.744 (0.651, 0.823) | 63.77 | 82.05 | 70.37 | 86.27 | 56.14 |
| Validation | 0.718 (0.564, 0.842) | 65.52 | 75.00 | 68.89 | 82.61 | 54.55 | |
| Medium | Training | 0.623 (0.525, 0.715) | 62.32 | 58.97 | 61.11 | 72.88 | 46.94 |
| Validation | 0.578 (0.421, 0.723) | 48.28 | 81.25 | 60.00 | 82.35 | 46.43 | |
| Slow | Training | 0.755 (0.663, 0.832) | 66.67 | 76.92 | 70.37 | 86.27 | 56.14 |
| Validation | 0.616 (0.460, 0.757) | 48.28 | 81.25 | 60.00 | 82.35 | 46.43 | |
| Combined | Training | 0.804 (0.716, 0.874) | 73.54 | 66.25 | 74.87 | 90.06 | 69.65 |
| Validation | 0.634 (0.477, 0.772) | 59.45 | 72.50 | 62.69 | 84.37 | 41.84 | |
| Whole lesion | Training | 0.597 (0.498, 0.690) | 17.39 | 100.00 | 47.22 | 100 | 40.63 |
| Validation | 0.571 (0.415, 0.718) | 72.41 | 56.25 | 66.67 | 75.00 | 52.94 |
ACC, accuracy; AUC, area under the receiver operating characteristic curve; CI, confidence interval; LR, logistic regression; NPV, negative predictive value; PPV, positive predictive value; SEN, sensitivity; SPE, specificity.