Table 2.

The AUC comparison of different ML models in different sets

ML models	Difference of AUC	S. E	95% CI	Z statistic	p
Train set
DT ~ KNN	0.0500	0.00551	0.0391—0.0608	9.059	< 0.0001
DT ~ SVM	0.0649	0.00629	0.0526—0.0772	10.321	< 0.0001
DT ~ XGB	0.0689	0.00433	0.0605—0.0774	15.940	< 0.0001
KNN ~ SVM	0.0150	0.00462	0.00591—0.0240	3.241	0.0012
KNN ~ XGB	0.0190	0.00356	0.0120—0.0260	5.339	< 0.0001
SVM ~ XGB	0.00404	0.00388	-0.00355—0.0116	1.043	0.2970
Test set
DT ~ KNN	0.0105	0.0267	-0.0418—0.0627	0.392	0.6952
DT ~ SVM	0.0528	0.0352	-0.0161—0.122	1.503	0.1328
DT ~ XGB	0.00503	0.0160	-0.0263—0.0363	0.315	0.7529
KNN ~ SVM	0.0424	0.0306	-0.0175—0.102	1.386	0.1656
KNN ~ XGB	0.0155	0.0219	-0.0275—0.0585	0.706	0.4801
SVM ~ XGB	0.0579	0.0317	-0.00421—0.120	1.827	0.0677
External validation set
DT ~ KNN	0.0882	0.0836	-0.0757—0.252	1.054	0.2917
DT ~ SVM	0.165	0.120	-0.0695—0.399	1.378	0.1681
DT ~ XGB	0.0368	0.0526	-0.0663—0.140	0.700	0.4838
KNN ~ SVM	0.0766	0.0978	-0.115—0.268	0.784	0.4332
KNN ~ XGB	0.125	0.0616	0.00434—0.246	2.030	0.0423
SVM ~ XGB	0.202	0.0789	0.0471—0.356	2.557	0.0106