Skip to main content
. 2023 Mar 29;22:44. doi: 10.1186/s12944-023-01808-0

Table 3.

Diagnostic performance of the nomogram, FLI, HSI, and AIP for predicting NAFLD in the training and validation sets

Models Training set (n = 1200) Validation set (n = 1118)
AUC (95%CI) P Youdan Sensitivity Specificity PPV NPV AUC (95% CI) P Youdan Sensitivity Specificity PPV NPV
Nomogram 0.863 (0.840–0.886) Ref 0.576 0.798 0.778 0.513 0.929 0.864 (0.841–0.887) Ref 0.583 0.922 0.661 0.448 0.966
FLI 0.862 (0.838–0.886) 0.850 0.564 0.787 0.777 0.508 0.926 0.866 (0.843–0.889) 0.772 0.600 0.851 0.748 0.502 0.944
HSI 0.835 (0.808–0.862) 0.019 0.548 0.812 0.736 0.474 0.931 0.833 (0.806–0.859) 0.006 0.514 0.817 0.697 0.446 0.927
AIP 0.782 (0.752–0.811)  < 0.001 0.444 0.761 0.683 0.413 0.907 0.773 (0.746–0.808)  < 0.001 0.444 0.767 0.677 0.415 0.907

AUROC area under the receiver operating characteristics, PPV positive predictive value, NPV negative predictive value, NAFLD nonalcoholic fatty liver disease, FLI fatty liver index, HSI hepatic steatosis index, AIP atherogenic index of plasma, Ref reference