Combining CT Images and Clinical Features of Four Periods to Predict Whether Patients Have Rectal Cancer

Table 6

The predictive performance of the prediction models using the training set.

Variables	Total model	BLS model
Cutoff	0.557	0.461
Sensitivity (95% CI)	0.970 (0.929–1.000)	0.985 (0.956–1.000)
Specificity (95% CI)	1.000 (1.000–1.000)	0.977 (0.951–1.000)
PPV (95% CI)	1.000 (1.000–1.000)	0.957 (0.908–1.000)
NPV (95% CI)	0.985 (0.964–1.000)	0.992 (0.977–1.000)
AUC (95% CI)	0.999 (0.996–1.000)	0.999 (0.997–1.000)
Accuracy (95% CI)	0.990 (0.976–1.000)	0.980 (0.960–0.999)

CI: confidence interval; PPV: positive predictive value; NPV: negative predictive value; AUC: area under the curve.