Predictive Model for Postoperative Delirium in Elderly Abdominal Cancer Patients Using SMOTE

A synthetic minority oversampling technique (SMOTE)-based model was developed to predict postoperative delirium in elderly patients with abdominal malignancies. In this cohort of 611 patients, risk factors for delirium included Charlson comorbidity index, American Society of Anesthesiologists classification, history of cerebrovascular disease, surgical duration, perioperative blood transfusion, and postoperative pain score. The SMOTE-based predictive model demonstrated comparable performance to traditional methods, with an area under the curve of 0.856, suggesting it effectively addresses data imbalance.