Table 3.

ROC curve analysis of three models in the training set.

Parameter	Logistic (Train)	SVM (Train)	Decision Tree (Train)
AUC	0.632 (CI: 0.533–0.730)	0.840 (CI: 0.653–0.758)	0.688 (CI: 0.601–0.775)
Accuracy	0.624 (CI: 0.530–0.707)	0.797 (CI: 0.719–0.862)	0.692 (CI: 0.606–0.769)
Sensitivity	0.654 (CI: 0.462–0.788)	0.846 (CI: 0.558–0.942)	0.558 (CI: 0.385–0.681)
Specificity	0.605 (CI: 0.272–0.741)	0.742 (CI: 0.284–0.852)	0.778 (CI: 0.575–0.904)

ROC, Receiver operating characteristic; AUC, area under the operating curve; CI, confidence interval.