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

Table 7

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

Variables	Total model	BLS model
Sensitivity (95% CI)	0.778 (0.586–0.970)	0.889 (0.744–1.000)
Specificity (95% CI)	1.000 (1.000–1.000)	0.906 (0.805–1.000)
PPV (95% CI)	1.000 (1.000–1.000)	0.842 (0.678–1.000)
NPV (95% CI)	0.889 (0.786–0.992)	0.935 (0.849–1.000)
AUC (95% CI)	0.962 (0.915–1.000)	0.965 (0.924–1.000)
Accuracy (95% CI)	0.920 (0.845–0.995)	0.900 (0.817–0.983)

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