Chen C, Yang D, Gao S, et al. - This retrospective research used machine learning methods to build a model to predict postoperative pneumonia in orthotopic liver transplantation (OLT) patients. From January 2015 to September 2019, data from 786 adult patients who had OLT at Sun Yat-sen University's Third Affiliated Hospital were retrospectively extracted from electronic medical records and randomly divided into a training set and a testing set. Five hundred ninety-one OLT patients were eventually involved and 253 (42.81%) were diagnosed with postoperative pneumonia, which was linked to increased postoperative hospitalization and mortality. Pneumonia was notably related to 14 items features: preoperative international normalized ratio, hematocrit, platelets, albumin, alanine transaminase, fibrinogen, white blood cell count, prothrombin time, serum sodium, total bilirubin, anesthesia time, preoperative length of stay, total fluid transfusion and operation time. The study showed that the XGBoost model with 14 common variables could predict postoperative pneumonia in OLT patients.