Application of machine learning in liver transplantation

Machine learning can efficiently extract the features and establish associations from complex databases, and better predict changes in clinical diseases by constructing models. Liver transplantation is one of the efficacious therapeutic options for all types of end-stage liver diseases and primary liver cancer. Nevertheless, it also faces multiple challenges. How to more effectively allocate the organs, expand the donor liver pool, evaluate the optimal donor-recipient matching, predict the complications after liver transplantation, disease recurrence and long-term survival have been the hot spots and difficulties. In recent years, certain progress has been made in the application of machine learning in the field of liver transplantation, showcasing promising prospect. In this article, the application status and prospect of machine learning in organ allocation before liver transplantation, donor liver evaluation, prediction of perioperative complications, blood transfusion, postoperative new disease, disease recurrence, acute rejection and long-term survival were reviewed, aiming to provide ideas and direction for subsequent investigations.