Table 4.

Performance evaluation of three models in three cohorts.

Model	Training cohort				Internal validation cohort				External validation cohort
	AUC (95% CI)	Sensitivity (%)	Specificity (%)	Accuracy (%)	AUC (95% CI)	Sensitivity (%)	Specificity (%)	Accuracy (%)	AUC (95% CI)	Sensitivity (%)	Specificity (%)	Accuracy (%)
Model1	0.820 (0.754–0.885)	61.80%	90.00%	74.22%	0.785 (0.663–0.906)	66.67%	86.67%	75.37%	0.808 (0.697–0.919)	66.67%	85.19%	75.00%
Model2	0.780 (0.707–0.853)	79.78%	68.57%	74.84%	0.794 (0.687–0.902)	66.67%	83.33%	73.91%	0.802 (0.690–0.915)	72.73%	81.48%	76.67%
Model3	0.874 (0.821–0.927)	64.05%	94.29%	77.36%	0.845 (0.744–0.946)	61.54%	100%	78.26%	0.841 (0.736–0.945)	87.88%	77.78%	83.34%

AUC, area under the curve; CI, confidence interval.