Table 3.
Area under the curve (AUC) and sensitivity and specificity at the optimal cutoff for DD based on all surfaces and by DD machine and tooth surface.
| Surface/Machine | AUC (95% CI) | Optimal DD cutoff | Sensitivity | Specificity |
|---|---|---|---|---|
| All surfaces | 0.911 (0.855–0.967) | 25 | 90.8% | 74.0% |
| Machine | ||||
| 1 | 0.926 (0.814–1.000) | 30 | 93.8% | 78.1% |
| 2 | 0.921 (0.826–1.000) | 30 | 89.9% | 76.7% |
| 3a | 0.941 (0.872–1.000) | 20 | 92.5% | 77.9% |
| Tooth surface | ||||
| Distal or mesial | 0.922 (0.836–1.000) | 20 | 98.4% | 75.8% |
| Facial or lingual | 0.923 (0.842–1.000) | 35 | 88.3% | 83.6% |
| Surface area | ||||
| Root | 0.954 (0.928–0.980) | 25 | 89.1% | 86.3% |
| Furcation and crown margin | 0.829 (0.685–0.972) | 45 | 67.9% | 78.3% |
| Tooth position | ||||
| Posterior | 0.899 (0.816–0.982) | 40 | 72.4% | 87.1% |
| Anterior | 0.958 (0.910–1.000) | 25 | 86.5% | 88.9% |
aDue the smaller sample sizes which produced training data sets with no variation in caries status (either all inactive or active caries) using tenfold cross-validation, fivefold cross-validation was used when the data were restricted to Machine 3.