Utilizing Artificial Intelligence for Predicting Postoperative Complications in Breast Reduction Surgery: A Comprehensive Retrospective Analysis of Predictive Features and Outcomes.

This article focuses on the development and evaluation of a machine learning (ML) model using gradient-boosting decision trees to predict severe complications within 30 days following breast reduction surgery. Based on a retrospective study of 322 cases at Tel Aviv Medical Center from 2017 to 2024, the model identified key predictive factors such as specimen weight, suprasternal notch-to-nipple (SN-N) distance, and liposuction volume, achieving an area under the receiver operating characteristic curve (AUC-ROC) of 0.83 and a negative predictive value (NPV) of 0.95. An interpretability tool was also created to help surgeons visualize complication risks based on clinical features, enhancing preoperative counseling and patient education. While the model shows promise for improving surgical planning and personalized care, the authors recommend further validation across diverse populations to confirm its broader clinical utility.