A Novel Diagnostic Nomogram for Noninvasive Evaluating Liver Fibrosis in Patients with Chronic Hepatitis B Virus Infection

Table 5

Comparing AUROC values of different models.

	AUROC	SE		95% CI for AUROC
				Lower	Upper
Training dataset
APRI scoring model	0.722	0.028	<0.001	0.668	0.777	31.02	<0.001
FIB-4 scoring model	0.729	0.027	<0.001	0.677	0.781	24.63	<0.001
APAG model	0.800	0.025	<0.001	0.752	0.848	7.79	0.005
S-index model	0.770	0.025	<0.001	0.721	0.819	13.21	<0.001
The nomogram	0.857	0.021	<0.001	0.816	0.899	—	—
Validation dataset
APRI scoring model	0.687	0.044	<0.001	0.600	0.773	19.96	<0.001
FIB-4 scoring model	0.749	0.042	<0.001	0.668	0.831	7.56	0.006
APAG model	0.791	0.038	<0.001	0.717	0.865	4.53	0.033
S-index model	0.772	0.039	<0.001	0.695	0.848	5.61	0.018
The nomogram	0.862	0.031	<0.001	0.801	0.922

The four serological diagnostic models are formulated as follows [4–7]: , , , , and .