machine learning–driven identification of determinants affecting clinical outcomes in outpatient surgery

The expanding use of artificial intelligence (ai) and machine learning (ml) within healthcare has created new avenues for forecasting clinical outcomes and improving operational efficiency. this study introduces a data-driven simulation framework designed to predict postoperative outcomes in outpatient surgeries through ml techniques, with a particular emphasis on the random forest algorithm. drawing on principles of process mining and predictive analytics in healthcare systems, the framework generates simulated patient-level datasets that incorporate key postoperative indicators, including length of stay, complication rates, and recovery trajectories. these datasets undergo extensive preprocessing, feature construction, and statistical verification to ensure reliability and representativeness. the model identified surgery duration and preoperative asa status as the most influential predictors, achieving strong accuracy with an r² of 0.81 and a mae of 0.29 days. the findings indicate that the framework effectively identifies major determinants of postoperative outcomes, such as anesthesia method, surgical duration, preoperative health conditions, and complication severity.