Table 3.

Predictive performance of variables in the training and validation sets.

Variables	Training set			Validation set
	AUC (95% CI)	Sensitivity (%)	Specificity (%)	AUC (95% CI)	Sensitivity (%)	Specificity (%)
Model-score	0.838 (0.731–0.914)	72.22%	88.68%	0.811 (0.634–0.927)	85.71%	76.00%
Rad-Score	0.800 (0.688–0.885)	66.67%	81.13%	0.794 (0.615–0.916)	100%	56.00%
Smoking	0.738 (0.621–0.836)	72.22%	75.47%	0.699 (0.481–0.824)	85.71%	48.00%
SUVmax	0.771 (0.656–0.862)	72.22%	73.58%	0.783 (0.602–0.908)	85.71%	72.00%