TABLE 6

Diagnostic performance of the models derived from the training cohort and applied to the training cohort (n=166)

AUC (95% CI)ThresholdSensitivity (95% CI)Specificity (95% CI)NPVPPV
Model 10.67 (0.59–0.76)0.200.78 (0.62–0.86)0.54 (0.35–0.67)0.590.74
Model 20.73 (0.65–0.81)0.220.80 (0.58–0.89)0.59 (0.41–0.71)0.630.77
Model 30.74 (0.65–0.82)0.330.75 (0.57–0.87)0.62 (0.44–0.74)0.590.77
Model 40.67 (0.58–0.76)0.240.74 (0.53–0.85)0.58 (0.36–0.71)0.590.73
Model 50.74 (0.67–0.83)0.120.85 (0.62–0.93)0.57 (0.41–0.69)0.690.77
Model 60.75 (0.67–0.83)0.140.83 (0.64–0.91)0.59 (0.34–0.71)0.690.76
Model 70.74 (0.65–0.82)0.340.73 (0.53–0.83)0.64 (0.42–0.76)0.600.76
Model 8#0.76 (0.68–0.84)0.250.77 (0.54–0.85)0.69 (0.44–0.81)0.660.80

AUC: area under the curve; NPV: negative predictive value; PPV: positive predictive value. #: best performing model.