Table 3 Diagnostic performance of the five models in training set and validation set.
Models | AUC (95%CI) | Sensitivity (95%CI) | Specificity (95%CI) | Precision (95%CI) | Accuracy (95%CI) |
|---|---|---|---|---|---|
The training set | |||||
 AIMeasure model | 0.735 (0.694–0.776) | 0.655 (0.652–0.658) | 0.667 (0.663–0.670) | 0.703 (0.700–0.706) | 0.660 (0.658–0.662) |
 BasicClinical model | 0.769 (0.731–0.808) | 0.697 (0.694–0.700) | 0.698 (0.695–0.702) | 0.735 (0.732–0.738) | 0.698 (0.696–0.699) |
 TotalClinical model | 0.826 (0.793–0.860) | 0.720 (0.717–0.723) | 0.714 (0.710–0.717) | 0.752 (0.749–0.755) | 0.717 (0.716–0.719) |
 AIBasicClinical model | 0.776 (0.738–0.814) | 0.700 (0.697–0.703) | 0.706 (0.702–0.709) | 0.741 (0.738–0.744) | 0.703 (0.701–0.704) |
 AITotalClinical model | 0.836 (0.804–0.869) | 0.733 (0.730–0.736) | 0.725 (0.722–0.729) | 0.763 (0.760–0.766) | 0.730 (0.728–0.731) |
The validation set | |||||
 AIMeasure model | 0.767 (0.707–0.827) | 0.636 (0.629–0.644) | 0.755 (0.747–0.762) | 0.757 (0.749–0.764) | 0.690 (0.686–0.694) |
 BasicClinical model | 0.781 (0.723–0.839) | 0.652 (0.644–0.659) | 0.709 (0.701–0.717) | 0.729 (0.721–0.736) | 0.678 (0.674–0.681) |
 TotalClinical model | 0.809 (0.754–0.863) | 0.682 (0.675–0.689) | 0.764 (0.756–0.771) | 0.776 (0.760–0.783) | 0.719 (0.715–0.723) |
 AIBasicClinical model | 0.781 (0.723–0.839) | 0.674 (0.667–0.681) | 0.718 (0.710–0.726) | 0.742 (0.735–0.749) | 0.694 (0.690–0.698) |
 AITotalClinical model | 0.818 (0.763–0.872) | 0.674 (0.667–0.681) | 0.773 (0.765–0.780) | 0.781 (0.774–0.788) | 0.719 (0.715–0.723) |
