Machine learning accurately predicts temporary stoma formation in CD patients

A machine learning model effectively predicts temporary stoma formation following intestinal resection in Crohn’s disease patients. An analysis of 252 patient records identified eight key predictors, achieving an area under the curve (AUC) between 0.886 and 0.998. The random forest algorithm showed the highest predictive accuracy, leading to potential improvements in surgical decision-making and personalized treatment approaches for patients. This innovative approach may enhance patient outcomes and streamline clinical processes for Crohn’s disease management.