Table 6.

The ROC curve of different models of LR-classifier in the training cohort.

Classifiers	Model	AUC	95% CI	Sensitivity	Specificity
LR	T1WI	0.795	0.663 - 0.893	0.565	0.806
	T2WI	0.951	0.855 - 0.991	0.826	0.864
	CE-T1WI	0.909	0.799 - 0.970	0.870f	0.871
	T1WI+T2WI	0.937	0.836 - 0.985	0.783	0.903
	T1WI+CE-T1WI	0.846	0.722 - 0.930	0.739	0.774
	T2WI+CE-T1WI	0.951	0.855 - 0.991	0.826	0.864
	T1WI+T2WI+CE-T1WI	0.940	0.840 - 0.986	0.913	0.903

T1WI, T1-weighted imaging; T2WI, T2-weighted imaging; AUC, area under curve; 95% CI, 95% confidence interval; LR, logistic regression.