July 12(4):453-460. doi: 10.1177/21501351211007106
Dimitris Bertsimas, Daisy Zhuo, Jack Dunn, Jordan Levine, Eugenio Zuccarelli, Nikos Smyrnakis, Zdzislaw Tobota, Bohdan Maruszewski, Jose Fragata, George E Sarris
Risk assessment tools typically used in congenital heart surgery (CHS) assume that various possible risk factors interact in a linear and additive fashion, an assumption that may not reflect reality. Using artificial intelligence techniques, we sought to develop nonlinear models for predicting outcomes in CHS.
We built machine learning (ML) models to predict mortality, postoperative mechanical ventilatory support time (MVST), and hospital length of stay (LOS) for patients who underwent CHS, based on data of more than 235,000 patients and 295,000 operations provided by the European Congenital Heart Surgeons Association Congenital Database. We used optimal classification trees (OCTs) methodology for its interpretability and accuracy, and compared to logistic regression and state-of-the-art ML methods (Random Forests, Gradient Boosting), reporting their area under the curve (AUC or c-statistic) for both training and testing data sets.
Optimal classification trees achieve outstanding performance across all three models (mortality AUC = 0.86, prolonged MVST AUC = 0.85, prolonged LOS AUC = 0.82), while being intuitively interpretable. The most significant predictors of mortality are procedure, age, and weight, followed by days since previous admission and any general preoperative patient risk factors.
The nonlinear ML-based models of OCTs are intuitively interpretable and provide superior predictive power. The associated risk calculator allows easy, accurate, and understandable estimation of individual patient risks, in the theoretical framework of the average performance of all centers represented in the database. This methodology has the potential to facilitate decision-making and resource optimization in CHS, enabling total quality management and precise benchmarking initiatives.
The authors present an excellent analysis exploring a novel risk-assessment tool for patients undergoing correction of a congenital heart defect. Utilizing artificial intelligence techniques, the authors attempted to develop nonlinear models for predicting outcomes in this patient population. Most prior models assumed a linear relationship between an event and adverse outcomes. The authors showed improved predictability of adverse outcomes when a non-linear estimated utilized. The authors should be congratulated with the innovativeness utilized to estimate these relationships. The ongoing struggle as to understand what outcomes from correction of a congenital heart defect should be. This is the result of several confounding issues, to include the vast case mix of these patients, the relative low incidence of congenital heart disease, and variations in both the operative and post-operative management of these individuals. This study provides us with another tool to assist with the complex and uncertain relationship.