Skip to main content

Table 6 Performance data of the neural network model for the classification of advanced and non-advanced thyroid carcinomas

From: Advanced thyroid carcinomas: neural network analysis of ultrasonographic characteristics

  Total
(n = 96)
ADV
(n = 32)
Non-ADV
(n = 64)
 
Accuracy (95% CI) 84.4 (75.5–91.0)    
Sensitivity (95% CI)   84.4 (67.2–94.7) 82.8 (71.3–91.1)  
Specificity (95% CI)   93.8 (84.8–98.3) 83.1 (71.7–91.2)  
Positive predictive value (95% CI)   87.1 (70.2–96.4) 82.8 (71.3–91.1)  
Negative predictive value (95% CI)   92.3 (83.0–97.5) 95.3 (88.4–98.7)  
Cohens Kappa (95% CI) 0.84 (71.2–97.5)    
F-score 0.84    
  1. ADV Advanced thyroid carcinoma (group 1), Non-ADV Non-advanced thyroid carcinoma (group 2 and 3), CI Confidence interval