Association Between Lactate-to-Albumin Ratio and 28-Day All-Cause Mortality in ICU-Admitted Acute Pancreatitis Patients: Development and Validation of a Machine Learning-Based Predictive Model Integra
3 hours ago
- #Machine Learning
- #Acute Pancreatitis
- #Predictive Model
- The lactate-to-albumin ratio (LAR) is a strong independent predictor of 28-day all-cause mortality in ICU-admitted acute pancreatitis patients, with the highest quartile (>1.72 mmol/g) showing significantly elevated risk (adjusted HR = 8.25).
- A machine learning-based predictive model was developed using three albumin-derived indices (LAR, red blood cell distribution width-to-albumin ratio [RAR], and blood urea nitrogen-to-albumin ratio [BAR]) along with clinical variables, with the Extratrees model demonstrating superior performance (AUC = 0.8546 in training, 0.852 in internal validation, and 0.82 in external validation).
- The study utilized data from the MIMIC-IV database (380 patients for training/internal validation) and the eICU-CRD database (298 patients for external validation), with predictors screened via the Boruta algorithm and model interpretability enhanced using SHAP.
- A web-based computational platform has been deployed for online individualized risk assessment, offering an efficient tool for precise risk stratification and early intervention in ICU-managed acute pancreatitis.