Utilizing Artificial Intelligence for Predicting Postoperative Complications in Breast Reduction Surgery: A...

Area of Science:

• Plastic Surgery
• Medical Informatics
• Machine Learning in Healthcare

Background:

• Breast reduction surgery is increasingly common for macromastia relief.
• Complication rates in breast reduction range widely from 6.2% to 43%.
• Predicting severe complications is crucial for patient safety and surgical planning.

Purpose of the Study:

• To develop and validate an AI model for predicting severe breast reduction complications.
• To identify key factors associated with severe complications post-surgery.
• To enhance preoperative counseling through risk visualization.

Main Methods:

• Retrospective analysis of 322 breast reduction cases (2017-2024).
• Development of a gradient-boosting decision tree machine learning model.
• Evaluation using 5-fold cross-validation, AUC-ROC, accuracy, sensitivity, and specificity.
• Utilized an interpretability tool for risk visualization.

Main Results:

• Severe complications occurred in 7.4% of patients.
• Predictive factors included specimen weight, SN-N distance, and liposuction volume.
• The AI model achieved an AUC-ROC of 0.83, accuracy of 0.93, and NPV of 0.95.
• The interpretability tool effectively visualized complication risks.

Conclusions:

• This is the first AI application for predicting severe complications in breast reduction.
• The developed AI model provides a reliable tool for surgical planning and patient education.
• Further validation in diverse populations is recommended to confirm clinical utility.