Potential value of CT-based comprehensive nomogram in predicting occult lymph node metastasis of esophageal squamous cell paralaryngeal nerves: a two-center study

Table 3 Diagnostic performance of different radiomics models for predicting osteoporosis in training and test cohorts

Model	Accuracy	AUC	95% CI	Sensitivity	Specificity	PPV	NPV
LR-train	0.83	0.78	0.71–0.85	0.37	0.97	0.78	0.84
LR-test	0.86	0.86	0.72–1.00	0.57	0.95	0.79	0.87
SVM-train	0.83	0.95	0.92–0.99	0.27	0.99	0.99	0.82
SVM-test	0.80	0.81	0.63–1.00	0.14	0.99	0.99	0.79
KNN-train	0.82	0.84	0.79–0.89	0.72	0.94	0.67	0.85
KNN-test	0.81	0.62	0.37–0.86	0.74	0.95	0.68	0.78
Random Forest-train	0.98	0.99	0.99–1.00	0.93	0.99	0.99	0.98
Random Forest-test	0.76	0.73	0.55–0.92	0.54	0.95	0.49	0.78
Light GBM-train	0.80	0.97	0.95–0.98	0.53	0.99	0.88	0.79
Light GBM-test	0.76	0.76	0.51–1.00	0.50	0.99	0.49	0.76
MLP-train	0.81	0.83	0.78–0.89	0.16	0.99	0.90	0.8
MLP-test	0.76	0.83	0.67–0.98	0.50	0.99	0.50	0.76

ISSN: 1479-5876