Table 3.
Accuracies, sensitivities, specificities, and area under the curve (AUC) values of prediction models in a leave-one-out cross-validation (LOOCV) of the primary dataset.
| Machine learning algorithm | Accuracy | Sensitivity | Specificity | AUC |
|---|---|---|---|---|
| LR | 0.834 | 0.833 | 0.836 | 0.915 |
| SVM | 0.866 | 0.902 | 0.800 | 0.932 |
| SNN | 0.796 | 0.833 | 0.727 | 0.896 |
| RF | 0.815 | 0.892 | 0.673 | 0.902 |
| NB | 0.809 | 0.853 | 0.727 | 0.867 |
| Mean ± SD | 0.824 ± 0.027 | 0.863 ± 0.033 | 0.753 ± 0.065 | 0.902 ± 0.024 |
| 95% CI | 0.790–0.858 | 0.822–0.903 | 0.672–0.833 | 0.873–0.932 |
LR: logistic regression, SVM: support vector machine, SNN: standard neural network, RF: random forest, NB: naïve Bayes.