Table 2. Five-fold cross-validation result comparison to other ML models with all features

	Matrix, mean (SD)
Model	Accuracy	Specificity	Sensitivity	Balanced accuracy	AUROC
XGB	0.6429(0.0050)	0.6421(0.0052)	0.7048(0.0151)	0.6734(0.0051)	0.7298(0.0051)
GBM	0.6590(0.0065)	0.6585(0.0067)	0.7007(0.0093)	0.6796(0.0051)	0.7388(0.0050)
LGBM	0.6542(0.0066)	0.6535(0.0066)	0.7069(0.0086)	0.6802(0.0055)	0.7401(0.0055)
Random forest	0.6232(0.0208)	0.6220(0.0212)	0.7191(0.0099)	0.6706(0.0064)	0.7284(0.0043)
AdaBoost	0.6434(0.0032)	0.6427(0.0033)	0.6982(0.0152)	0.6705(0.0077)	0.7259(0.0065)
LR	0.6460(0.0017)	0.6448(0.0019)	0.7363(0.0175)	0.6906(0.0082)	0.7538(0.0091)
GBM + LGBM	0.6565(0.0052)	0.6559(0.0053)	0.7043(0.0080)	0.6801(0.0050)	0.7395(0.0052)
GBM + LR	0.6952(0.0038)	0.6953(0.0039)	0.6857(0.0178)	0.6905(0.009)	0.7529(0.0079)
LGBM + LR	0.6823(0.0038)	0.682(0.0039)	0.6996(0.0164)	0.6908(0.0081)	0.7530(0.0080)
GBM + LGBM + LR	0.6892(0.0046)	0.6893(0.0047)	0.6879(0.0152)	0.6886(0.0076)	0.7503(0.0072)

Abbreviations: SD, standard deviation; AUROC, area under receiver of characteristics; XGB, XGBoost; GBM, gradient boosting machine; LGBM, light gradient-boosting machine; AdaBoost, adaptive boosting; LR, logistic regression.
The bold characters were the best-performance model.