Early Prediction of Liver Cirrhosis Up to Two Years in Advance: A Machine Learning Study Benchmarking Against the FIB-4 and APRI Scores
Researchers have developed machine learning models capable of predicting liver cirrhosis up to two years before diagnosis using electronic health record data. These models, particularly an XGBoost implementation, demonstrated superior performance compared to traditional clinical scores like FIB-4 and APRI. The study suggests these ML tools could be integrated into clinical workflows to enable earlier risk stratification and proactive patient management. AI
IMPACT Early detection of liver cirrhosis via ML could improve patient outcomes and reduce healthcare costs.