Table 5 Performance data of the neural network model for the classification of the study groups
From: Advanced thyroid carcinomas: neural network analysis of ultrasonographic characteristics
Total (n = 96) | ADV (n = 32) | PTC (n = 32) | FTC (n = 32) | |
|---|---|---|---|---|
Accuracy (95% CI) | 84.4 (75.5–91.0) | |||
Sensitivity (95% CI) | 84.4 (67.2–94.7) | 87.5 (71.0–96.5) | 78.1 (60.0–90.7) | |
Specificity (95% CI) | 93.8 (84.8–98.3) | 92.2 (82.7–97.4) | 90.6 (79.7–96.5) | |
Positive predictive value (95% CI) | 87.1 (70.2–96.4) | 84.9 (68.1–94.9) | 80.6 (62.5–92.6) | |
Negative predictive value (95% CI) | 92.3 (83.0–97.5) | 93.6 (84.5–98.2) | 89.2 (79.1–95.6) | |
Fleiss Kappa (95% CI) | 0.56 (0.55–0.57) | |||
F-score | 0.84 |