Construction of a Diagnostic Model for Lymph Node Metastasis of the Papillary Thyroid Carcinoma Using Preoperative Ultrasound Features and Imaging Omics
Table 3
The predictive performance of these models in the training and testing sets.
Models
Cut-off
Sensitivity (95% CI)
Specificity (95% CI)
PPV (95% CI)
NPV (95% CI)
AUC (95% CI)
Accuracy (95% CI)
Model 1a
0.348
0.849 (0.806–0.891)
0.851 (0.817–0.886)
0.790 (0.744–0.837)
0.895 (0.864–0.925)
0.913 (0.893–0.934)
0.850 (0.823–0.877)
Model 1b
0.348
0.738 (0.661–0.815)
0.771 (0.707–0.835)
0.710 (0.632–0.788)
0.795 (0.733–0.857)
0.813 (0.762–0.863)
0.757 (0.708–0.806)
Model 2a
0.437
0.808 (0.761–0.855)
0.868 (0.836–0.901)
0.802 (0.755–0.849)
0.873 (0.840–0.905)
0.913 (0.892–0.934)
0.844 (0.817–0.872)
Model 2b
0.437
0.730 (0.653–0.808)
0.789 (0.727–0.851)
0.724 (0.647–0.802)
0.794 (0.732–0.856)
0.818 (0.769–0.868)
0.764 (0.715–0.812)
Model 3a
0.360
0.886 (0.848–0.924)
0.827 (0.790–0.863)
0.772 (0.725–0.818)
0.916 (0.888–0.944)
0.941 (0.925–0.957)
0.850 (0.823–0.877)
Model 3b
0.360
0.762 (0.688–0.836)
0.723 (0.655–0.791)
0.676 (0.599–0.753)
0.800 (0.736–0.864)
0.821 (0.772–0.871)
0.740 (0.689–0.790)
Model 4a
0.500
0.823 (0.777–0.868)
0.966 (0.948–0.983)
0.941 (0.911–0.971)
0.892 (0.863–0.921)
0.984 (0.977–0.990)
0.909 (0.887–0.931)
Model 4b
0.500
0.667 (0.584–0.749)
0.910 (0.866–0.953)
0.848 (0.778–0.919)
0.782 (0.724–0.841)
0.857 (0.811–0.902)
0.805 (0.759–0.850)
aUsing the training set; busing the testing set. PPV: positive predictive value; NPV: predictive value; AUC: area under the curve; CI: confidence internal.
