Minireviews
Copyright ©The Author(s) 2025.
Artif Intell Gastroenterol. Jun 8, 2025; 6(1): 107193
Published online Jun 8, 2025. doi: 10.35712/aig.v6.i1.107193
Table 2 Comparative analysis of artificial intelligence models in alcohol-related liver disease prediction
AI model
Methodology
Performance metrics
Key limitations
Citation
Gradient BoostingMICE imputation, SMOTE, feature selectionAUC = 0.87 (30-day mortality prediction)Small sample size, Lack of external validation, missing dataGao et al[2], 2022
Stacked Ensemble (XGBoost + Logistic Regression)Multi-omics + clinical featuresAccuracy = 93.86% (alcoholic cirrhosis prediction)Lack of external validationVinutha et al[6], 2022
Random Forest/XGBoostBayesian optimizationAccuracy = 81.06% (Random Forest), 79.85% (XGBoost)Data heterogeneityKumar and Rani[7], 2024
Extra Tree with OversamplingLiver stiffness + clinical dataAccuracy = 92% (early ARLD detection)Requires high-resolution imagingLima et al[10], 2024