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

BMC Oral Health

Table 4 Diagnostic performance of the framework and individual dentists for missing teeth

Var	Spe	Sen	Youden’s index	AUC (SE; 95% CI)	P	Z
Framework	0.994	0.885	0.879	0.939 (0.00708; 0.934–0.944)	ref	ref
H1	0.995	0.889	0.884	0.942 (0.00697; 0.937–0.947)	0.748	0.321
H2	0.998	0.914	0.912	0.956 (0.00620; 0.951–0.960)	0.023*	2.279
H3	0.997	0.920	0.917	0.958 (0.00601; 0.954–0.963)	0.012*	2.515
M1	0.997	0.885	0.882	0.941 (0.00707; 0.935–0.946)	0.857	0.180
M2	0.996	0.900	0.896	0.948 (0.00663; 0.943–0.953)	0.268	1.109
M3	0.996	0.869	0.865	0.933 (0.00747; 0.927–0.938)	0.448	0.759
L1	0.994	0.832	0.826	0.913 (0.00828; 0.907–0.919)	0.006*	2.770
L2	0.997	0.875	0.872	0.936 (0.00732; 0.931–0.941)	0.707	0.376
L3	0.996	0.852	0.848	0.924 (0.00787; 0.918–0.930)	0.053	1.938

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