Table 5.

Predictive ability of the radiomics nomogram, radiomics signature, and clinical model for the classification of low and high Ki-67 lung cancer.

Variable		AUC	(95% CI)	Accuracy	Sensitivity	Specificity
Clinical model	Train	0.77	(0.69~0.85)	0.69	0.59	0.82
	Test	0.72	(0.62~0.83)	0.66	0.54	0.80
Radiomics signature	Train	0.88	(0.82~0.93)	0.81	0.79	0.84
	Test	0.86	(0.78~0.94)	0.80	0.75	0.88
Radiomics nomogram	Train	0.91	(0.85~0.98)	0.83	0.79	0.85
	Test	0.87	(0.80~0.95)	0.83	0.75	0.90