Skip to main content
Insights into Imaging

Table 4 Diagnostic performance of models for predicting histological differentiation of hepatocellular carcinoma

From: Predicting histologic differentiation of solitary hepatocellular carcinoma up to 5 cm on gadoxetate disodium-enhanced MRI

 

Training set

Validation set

AUC (95%CI)

0.802 (0.703–0.900)

0.830 (0.693–0.966)

Sensitivity

66.7% (20/30)

66.7% (8/12)

Specificity

85.7% (84/98)

81.0% (34/42)

PPV

58.8% (20/34)

50.00% (8/16)

NPV

89.4% (84/94)

89.5% (34/38)

ACC

81.3% (104/128)

77.8% (42/54)

  1. Numbers in parentheses are 95% CIs, and numbers in brackets are raw data
  2. AUC, receiver operating characteristic curve; PPV, positive predictive value; NPV, negative predictive value; ACC, accuracy
Back to article page