Using Machine Learning Algorithms to Predict High-Risk Factors for Postoperative Delirium in Elderly Patients

Postoperative delirium (POD) is a common postoperative complication in elderly patients, and it greatly affects the short-term and long-term prognosis of patients. The purpose of this study was to develop a machine learning model to identify preoperative, intraoperative and postoperative high-risk factors and predict the occurrence of delirium after nonbrain surgery in elderly patients.

A total of 950 elderly patients were included in the study, including 132 patients with POD. We collected 30 characteristic variables, including patient demographic characteristics, basic medical history, preoperative examination characteristics, type of surgery, and intraoperative information. Three machine learning algorithms, multilayer perceptron (MLP), extreme gradient boosting (XGBoost), and k-nearest neighbor algorithm (KNN), were applied to construct the model, and the k-fold cross-validation method, ROC curve, calibration curve, decision curve analysis (DCA) and external validation were used for model evaluation.

XGBoost showed the best performance among the three prediction models. The ROC curve results showed that XGBoost had a high area under the curve (AUC) value of 0.982 in the training set; the AUC value in the validation set was 0.924, and the prediction model was highly accurate. The k-fold cross-validation method was used for internal validation, and the XGBoost model was stable The calibration curve showed high predictive power of the XGBoost model. The DCA curve showed a higher benefit rate for patients who received interventional treatment under the XGBoost model. The AUC value for the external validation set was 0.88, indicating that the predictive model was extrapolative.

The prediction model of POD derived from the machine learning algorithm in this study has high prediction accuracy and clinical utility, which is beneficial for clinicians to diagnose and treat patients in a timely manner.