A Machine Learning Architecture to Predict Post-Hepatectomy Liver Failure Using Liver Regeneration Biomarkers and Time-Phased Data

Post-hepatectomy liver failure (PHLF) is the leading cause of morbidity and mortality following major hepatectomy. Existing prediction models fail to capture the dynamic liver regeneration and perioperative changes, limiting their predictive accuracy. We aimed to develop a machine learning (ML) modelling system (PILOT architecture) integrating liver regeneration biomarkers with time-phased perioperative clinical data to accurately predict PHLF risk.