Table 3.

Predictive performance of the clinical model, radiomics score, and radiomics nomogram for high-grade endometrial cancer.

Cohort	Models	AUC (95%CI)	ACC	SEN	SPE	NPV	PPV
Training set	Clinical model	0.837 (0.754, 0.920)	0.801	0.714	0.846	0.846	0.714
	Radiomics score	0.875 (0.797, 0.952)	0.830	0.889	0.808	0.952	0.632
	Radiomics nomogram	0.923 (0.869, 0.977)	0.877	0.741	0.918	0.905	0.769
Validation set	Clinical model	0.857 (0.741, 0.973)	0.721	1.000	0.657	1.000	0.400
	Radiomics score	0.786 (0.592, 0.979)	0.837	0.625	0.886	0.912	0.556
	Radiomics nomogram	0.914 (0.827, 0.998)	0.839	1.000	0.771	1.000	0.500

AUC, area under curve; ACC, accuracy; SEN, sensitivity; SPE, specificity; PPV, positive predictive value, NPV, negative predictive value.