Artificial intelligence in the diagnosis of dental diseases on panoramic radiographs: a preliminary study

BMC Oral Health

Table 3 Diagnostic performance of the framework and individual dentists for full crowns

Var	Spe	Sen	Youden’s index	AUC (SE; 95% CI)	P	Z
Framework	0.998	0.953	0.951	0.975 (0.00641; 0.972–0.978)	ref	ref
H1	0.999	0.971	0.970	0.985 (0.00508; 0.982–0.988)	0.241	1.173
H2	0.999	0.960	0.959	0.980 (0.00592; 0.976–0.982)	0.642	0.465
H3	1.000	0.964	0.964	0.982 (0.00566; 0.979–0.984)	0.458	0.743
M1	0.999	0.953	0.952	0.976 (0.00641; 0.973–0.979)	0.944	0.070
M2	1.000	0.938	0.938	0.969 (0.00728; 0.965–0.972)	0.464	0.732
M3	1.000	0.953	0.953	0.976 (0.00641; 0.973–0.979)	0.922	0.098
L1	0.999	0.920	0.919	0.960 (0.00820; 0.955–0.964)	0.130	1.516
L2	0.999	0.909	0.908	0.954 (0.00869; 0.950–0.958)	0.037*	2.090
L3	0.999	0.956	0.955	0.978 (0.00617; 0.974–0.981)	0.775	0.286

Delong test was conducted for statistically analysis of the difference of AUC between the framework and the dentists. * represents significant difference (p < 0.05)

ISSN: 1472-6831