Table 4.
Comparison of the three predictive models in the test set.
| Model | Area under the curve (%) | Sensitivity (%) | Specificity (%) | Accuracy (%) | Positive predictive value (%) | Negative predictive value (%) | Composite Index |
| Logistic regression | 68.1 | 58.62 | 77.53 | 75.91 | 19.62 | 95.24 | 0.00 |
| Support vector machine | 87.5 | 88.51 | 99.78 | 98.82 | 97.47 | 98.93 | 1.00 |
| Random forest | 79.6 | 83.91 | 98.71 | 97.44 | 85.88 | 98.5 | 0.81 |