Observational Study
Copyright ©The Author(s) 2025.
World J Gastroenterol. Jul 28, 2025; 31(28): 108321
Published online Jul 28, 2025. doi: 10.3748/wjg.v31.i28.108321
Table 3 Diagnostic performances of the machine learning models based on the BIU index for the diagnosis of ≥ F2 in the training and validation sets

AUC (95%CI)
P value1
Sensitivity
Specificity
Accuracy
PPV
NPV
Kappa
Training set
    LR0.81 (0.71, 0.90)Reference0.840.740.800.810.790.59
    DT0.89 (0.82, 0.95)0.5470.800.830.810.860.760.62
    SVM0.98 (0.95, 1.00)< 0.0010.910.910.910.930.890.82
    RF1.00 (1.00, 1.00)< 0.0011.001.001.001.001.001.00
    ANN0.82 (0.73, 0.91)0.5340.840.800.830.840.800.64
    XGBoost1.00 (0.99, 1.00)< 0.0010.981.000.991.000.970.98
Internal validation set
    LR0.88 (0.70, 1.00)Reference0.820.700.760.750.780.52
    DT0.80 (0.61, 0.96)0.1870.730.800.760.800.730.53
    SVM0.95 (0.84, 1.00)0.3770.910.600.760.710.860.52
    RF0.86 (0.67, 0.99)0.8411.000.600.810.731.000.61
    ANN0.90 (0.76, 1.00)0.7620.910.800.860.830.890.71
    XGBoost0.75 (0.50, 0.95)0.3250.820.600.710.690.750.42