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

Table 4

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

Variables	Clinical demographics	Radiomic	Total model
Cut off	0.426	0.265	0.557
Sensitivity (95% CI)	0.910 (0.842–0.979)	0.970 (0.929–1.000)	0.970 (0.929–1.000)
Specificity (95% CI)	0.884 (0.828–0.939)	0.907 (0.857–0.957)	1.000 (1.000–1.000)
PPV (95% CI)	0.803 (0.713–0.892)	0.844 (0.763–0.925)	1.000 (1.000–1.000)
NPV (95% CI)	0.950 (0.911–0.989)	0.983 (0.960–1.000)	0.985 (0.964–1.000)
AUC (95% CI)	0.938 (0.903–0.972)	0.980 (0.966–0.994)	0.999 (0.996–1.000)
Accuracy (95% CI)	0.893 (0.850–0.936)	0.929 (0.893–0.965)	0.990 (0.976–1.000)

Compared with the total model, the difference is statistically significant. CI: confidence interval; PPV: positive predictive value; NPV: negative predictive value; AUC: area under the curve.