Table 4.

Model performances (based on evaluation metrics) on the testing set.

Model name	Accuracy	Precision	Recall	F1-score	Specificity	AUROCa
LRb	0.750	0.360	0.720	0.540	0.750	0.832
SVMc	0.860	1.000	0.130	0.565	1.000	0.920
KNNd	0.910	0.660	0.910	0.785	0.910	0.938
Decision tree	0.877	0.624	0.605	0.615	0.930	0.767
Random forest	0.923	0.777	0.733	0.755	0.959	0.962
XGBooste	0.930	0.850	0.720	0.785	0.970	0.960
AdaBoostf	0.900	0.700	0.650	0.675	0.950	0.900
DNNg	0.927	0.790	0.746	0.768	0.962	0.933

^aAUROC: area under the receiver operating characteristic curve.

^bLR: logistic regression.

^cSVM: support vector machine.

^dKNN: k-nearest neighbor.

^eXGBoost: extreme gradient boost.

^fAdaBoost: adaptive boosting.

^gDNN: deep neural network.