Machine learning model predicts unplanned reoperation risk in rectal cancer

An optimized machine learning model was developed to predict the risk of unplanned reoperation (uro) after anterior resection for rectal cancer, utilizing data from 2,384 patients. The Random Forest model showed exceptional performance, achieving an accuracy of 84.2% and an AUC of 0.889. Key factors influencing uro risk included tumor location, previous surgeries, and operative time. An online platform for real-time risk assessment has been created, potentially aiding clinical decision-making and enhancing patient outcomes.