Skip to main content

Table 6 Diagnostic performance of the framework and individual dentists for caries

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

Var

Spe

Sen

Youden’s index

AUC (SE; 95% CI)

P

Z

Framework

0.990

0.554

0.544

0.772 (0.00949; 0.764–0.781)

ref

ref

H1

0.980

0.621

0.601

0.800 (0.00928; 0.792–0.809)

0.007*

2.721

H2

0.944

0.755

0.699

0.849 (0.00830; 0.842–0.857)

< 0.0001*

7.099

H3

0.975

0.669

0.644

0.822 (0.00901;0.814–0.830)

< 0.0001*

4.799

M1

0.978

0.569

0.547

0.773 (0.00947; 0.765–0.782)

0.917

0.104

M2

0.976

0.586

0.562

0.781 (0.00943;0.772–0.789)

0.424

0.800

M3

0.988

0.546

0.534

0.767 (0.00951; 0.758–0.775)

0.623

0.492

L1

0.988

0.417

0.405

0.702 (0.00942; 0.693–0.712)

< 0.0001*

6.847

L2

0.995

0.509

0.504

0.752 (0.00954; 0.743–0.761)

0.060

1.884

L3

0.990

0.402

0.392

0.696 (0.00936; 0.687–0.706)

< 0.0001*

6.859

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

BMC Oral Health

ISSN: 1472-6831

Contact us