Table 3 The nomogram comparing with other models.

From: Noninvasive diagnosis of significant liver fibrosis in patients with chronic hepatitis B using nomogram and machine learning models

Models

Cut-off

Training set

 

Validation set

AUC (95%CI)

Sensitivity

Specificity

NPV

PPV

AUC (95%CI)

Sensitivity

Specificity

NPV

PPV

nomogram

0.410

0.806(0.740, 0.872)

0.696

0.788

0.804

0.640

0.808(0.707, 0.909)

0.625

0.756

0.739

0.645

APRI

1.676

0.573(0.488, 0.658)

0.217

0.912

0.656

0.600

0.625(0.501, 0.749)

0.125

0.911

0.594

0.500

FIB-4

1.360

0.639(0.554, 0.724)

0.478

0.788

0.712

0.579

0.710(0.592, 0.829)

0.406

0.867

0.672

0.684

GPR

0.613

0.649(0.565, 0.734)

0.420

0.858

0.708

0.644

0.716(0.602, 0.830)

0.344

0.889

0.656

0.688

  1. NPV: negative predictive value; PPV: positive predictive value.