A novel super learner machine learning model significantly enhances the predictive accuracy of postoperative complications in colorectal surgery. Analyzing data from over 14,000 cases, the model exceeded traditional logistic regression and extreme gradient boosting methods, demonstrating an area under the receiver operating characteristic curve (AUROC) surpassing 0.94 for mortality predictions. This advancement promises to improve personalized patient counseling, refine clinical decision-making, and optimize healthcare resources in surgical settings.
Journal Article by Violante T, Ferrari D (…) Larson DW et 4 al. in Ann Surg
Copyright © 2025 Wolters Kluwer Health, Inc. All rights reserved.