Development and validation of machine learning models for predicting acute kidney injury in acute-on-chronic liver failure: a multimodel comparative study - PubMed
3 hours ago
- #acute kidney injury
- #liver failure
- #machine learning
- Study aimed to develop machine learning models to predict acute kidney injury (AKI) in acute-on-chronic liver failure (ACLF) patients.
- Retrospective study included 1,076 ACLF patients, with 23.2% developing AKI during hospitalization.
- Six machine learning models were evaluated: logistic regression, random forest, k-nearest neighbors, support vector machine, decision tree, and XGBoost.
- Random forest model performed best with AUC-ROC of 0.899 and AUC-PR of 0.806.
- Key predictors included age, hypertension, total bilirubin, blood urea nitrogen, serum creatinine, and other clinical factors.
- Findings suggest ML models, especially random forest, can effectively identify high-risk AKI patients in ACLF.