The development group was used for model development, five-fold cross-validation, and model optimization. The investigators collected a comprehensive set of variables for feature selection, including preoperative demographic characteristics (sex, age, body mass index, marital status, occupation, etc.), laboratory indicators (routine blood and urine tests, liver and kidney function, coagulation function, etc.), preoperative comorbidities, and surgical information (surgical department, surgical classification, American Society of Anesthesiologists physical status classification, intraoperative position, fluid intake and output, vital signs, intraoperative medication use, etc.). Subsequently, multiple machine learning methods, including logistic regression, extreme gradient boosting, decision tree, random forest, and Bayesian approaches, were employed for model building and optimization.

The external (time) validation group is used for future generalization ability assessment. The investigators prospectively collected patient-related data. In addition to the same variables as those in the development group and the testing group, the investigators also evaluated and collected the frailty status of patients before the operation, and recorded prognostic indicators such as the incidence of in-hospital complications, in-hospital mortality, length of hospital stay and hospitalization cost of patients. The investigators used the data from the external (time) validation group to validate the model performance, incorporated the frailty assessment as a new predictor into the model, calculated the incremental values and evaluated the performance of the updated model.

The exposure factors were the perioperative related operations experienced by the patients and their individual conditions