Artificial Intelligence in Liver Transplantation: Clinical Applications, Challenges, and Future Directions

Liver transplantation remains a technically demanding procedure despite continued surgical advances. Successful outcomes depend on balancing donor selection with perioperative complexity, where each decision shapes graft and patient survival. Conventional scoring systems such as Model for End-Stage Liver Disease (MELD)-based prioritisation only partially explain differences in patient outcomes across transplant centres worldwide. Advances in artificial intelligence (AI) and machine learning (ML)-including longitudinal modelling of electronic health records, deep learning for imaging, and multimodal data integration-offer new opportunities to individualise risk stratification for graft selection, predict allograft dysfunction and rejection, and improve long-term survival. Clinically developed AI models may strengthen organ matching decisions and acceptance, while identifying candidates who require intensified immunosuppressant monitoring or targeted diagnostic testing in the postoperative period, without increasing rejection risk. However, introducing AI/ML systems requires thorough validation, careful calibration, and ongoing clinical audit to reduce bias, detect technical errors, and ensure reproducible performance. This review summarises current and emerging clinical applications of AI in liver transplantation and explores near-future directions. These include AI-driven graft evaluation during organ retrieval and normothermic machine perfusion, along with emerging evidence for the safety, reliability, and clinical utility of AI in transplantation practice.