Table 2.
The performance of SurgHipNet classification results in the testing data set.
Features | Values |
True positive | 92 |
True negative | 372 |
False positive | 3 |
False negative | 8 |
Accuracy | 0.977 |
Prevalence | 0.211 |
Sensitivity | 0.920 |
Specificity | 0.992 |
Positive predictive value | 0.968 |
Negative predictive value | 0.979 |
F1-score | 0.944 |
Area under the receiver operating characteristic curve (95% CI) | 0.994 (0.990-0.998) |