Table 3.
Model prediction performance in the validation set.
| Algorithm | AUC (95%CI) | Accuracy | Sensitivity | Specificity | PPV | NPV | F1 Score |
|---|---|---|---|---|---|---|---|
| SVM | 0.912 (0.827–0.998) | 0.845 | 0.846 | 0.842 | 0.917 | 0.727 | 0.880 |
| Adaboost | 0.881 (0.777–0.986) | 0.897 | 0.897 | 0.895 | 0.946 | 0.810 | 0.921 |
| XGBoost | 0.871 (0.760–0.982) | 0.879 | 0.872 | 0.895 | 0.944 | 0.773 | 0.907 |
| RF | 0.895 (0.788–1.000) | 0.897 | 0.897 | 0.895 | 0.946 | 0.810 | 0.921 |
| KNN | 0.864 (0.755–0.973) | 0.862 | 0.872 | 0.842 | 0.919 | 0.762 | 0.895 |
| LR | 0.899 (0.788–1.000) | 0.776 | 1.000 | 0.316 | 0.750 | 1.000 | 0.857 |
AUC = Area Under Curve, SVM = Support Vector Machine, AdaBoost = Adaptive Boosting, XGBoost = eXtreme Gradient Boosting, RF = Random Forest, KNN = K-Nearest Neighbors, LR = Logistic Regression, PPV=Positive Predictive Value, NPV=Negative Predictive Value.