Skip to main content

Table 5 Diagnostic performance of LI-RADS categories in the discrimination of HCCs from benign nodules

From: Differentiation of small (≤ 3 cm) hepatocellular carcinomas from benign nodules in cirrhotic liver: the added additive value of MRI-based radiomics analysis to LI-RADS version 2018 algorithm

Diagnostic pattern

Az (95%CI)

Sensitivity

Specificity

PPV

NPV

Accuracy

LI-RADS

0.898 (0.834–0.961)

93.8% (105/112)

81.8% (36/44)

92.9% (105/113)

83.7% (36/43)

90.4% (141/156)

Radiomics signature

0.917 (0.860–0.974)

93.8% (105/112)

86.4% (38/44)

94.6% (105/111)

84.4% (38/45)

91.7% (143/156)

Combined nomogram

0.975 (0.954–0.996)

97.3% (109/112)

97.7% (43/44)

99.1% (109/110)

93.5% (43/46)

97.4% (152/156)

  1. LI-RADS liver imaging reporting and data system, Az area under the receiver operating characteristic curve, PPV positive predictive value, NPV negative predictive value