Personalizing Patient Risk of a Life Altering Event: An Application of Machine Learning to Hemiarch Surgery

Objective To assess a machine learning model’s ability to predict the occurrence of life altering events (LAE) in hemiarch surgery and determine contributing patient characteristics and intraoperative factors. Methods In total, 602 patients who underwent hemiarch replacement at a high-volume, aortic center from 2009-2022 were included. Patients were randomly divided into training (80%) and testing (20%) sets with various eXtreme gradient boosting (XGBoost) candidate models constructed to predict the risk of experiencing LAE, including stroke, mortality, or new renal replacement therapy requirement. 64 input parameters from the index hospitalization were identified, including 24 demographic characteristics as well as 8 pre-operative and 32 intra-operative variables. A SHapley Additive exPlanation (SHAP) beeswarm plot was generated to identify and interpret the impact of individual features on the predictions of the final model. Results A LAE was noted in 15% (90/602) of patients who underwent hemiarch replacement, including urgent/emergent cases and dissections. The final XGBoost model demonstrated a cross-validation accuracy of 88% on the testing set and was well-calibrated as evidenced by a low Brier score of 0.12. The best performing model achieved an area under the receiver-operating characteristic curve of 0.76 and an area under the precision-recall curve of 0.55. The SHAP beeswarm plot provided insights into key features that significantly influenced model prediction. Conclusions Machine learning demonstrated superior accuracy in predicting hemiarch patients that would experience a LAE. This model may help to guide patients and clinicians in stratifying risk on an individual basis, which may in turn influence clinical decision-making.