Table 4.
Discrimination and calibration of the three imaging models.
| Model | AUC (95% CI) | Sensitivity (%)a | Specificity (%)a | PPV (%) | NPV (%) | HL χ2 (P) | Brier |
|---|---|---|---|---|---|---|---|
| Radiographic | 0.696 (0.638–0.754) | 27.3 | 98.3 | 77.4 | 86.6 | 1.17 (0.997) | 0.116 |
| Ultrasound | 0.842 (0.797–0.887) | 63.6 | 91.2 | 60.2 | 92.3 | 1.70 (0.989) | 0.091 |
| Combined | 0.862 (0.821–0.903) | 65.9 | 93.8 | 69.0 | 92.9 | 4.39 (0.820) | 0.075 |
At the cut-off of 0.3; PPV, positive predictive value; NPV, negative predictive value; HL, Hosmer–Lemeshow goodness-of-fit test.