Related Experiment Video
Updated: Jun 11, 2026
Estimating Bilateral Atrial Function by Cardiovascular Magnetic Resonance Feature Tracking in Patients with Paroxysmal Atrial Fibrillation
Published on: July 20, 2022
Machine learning-based prediction of postoperative atrial fibrillation risk in coronary artery bypass grafting
Yang Zhang1, Zhihan Zhang1, Hu Zhang1
1Department of Cardiovascular Surgery, The Affiliated Hospital of Xuzhou Medical University, 99 Huaihai West Road, Xuzhou, 221000, China.
Insights
A new stacking machine learning model shows moderate accuracy in predicting postoperative atrial fibrillation (POAF) after coronary artery bypass grafting (CABG). The model
Area of Science:
- Cardiovascular Surgery
- Medical Informatics
- Machine Learning in Healthcare
Background:
- Postoperative atrial fibrillation (POAF) complicates 20-40% of coronary artery bypass grafting (CABG) procedures, increasing patient morbidity and mortality.
- Existing machine learning (ML) models for POAF prediction are limited by single-center data, small sample sizes, and the use of single algorithms.
Purpose of the Study:
- To develop and internally validate a stacking ensemble ML model for predicting POAF in CABG patients.
- To identify key predictors of POAF and assess the model's incremental clinical value compared to traditional methods.
Main Methods:
- Retrospective analysis of 563 CABG patients, with data split into training (n=394) and validation (n=169) sets.
- Feature selection using elastic net with stability selection; development of nine base ML algorithms and a stacking ensemble.
- Model performance evaluated using discrimination (AUC), calibration, and decision curve analysis (DCA).
Main Results:
- Thirteen predictors were identified, with age, intraoperative phenylephrine use, and stroke history being most significant.
- The stacking model achieved a validation AUC of 0.7425 and an F1 score of 0.711.
- The stacking model demonstrated superior performance over logistic regression and clinical risk scores, though DCA indicated limited net clinical benefit.
Conclusions:
- The developed stacking model offers moderate discrimination for POAF prediction post-CABG but is not yet clinically ready without external validation.
- Identified predictors warrant further investigation for mechanistic studies, but their clinical utility requires further proof.
- The model's net clinical benefit is confined to a narrow threshold range, limiting its immediate applicability.
Background:
Postoperative atrial fibrillation (POAF) occurs in 20-40% of patients undergoing coronary artery bypass grafting (CABG) and is associated with increased morbidity and mortality. Existing machine learning (ML) studies are limited by single‑center designs, small samples, and reliance on single algorithms.
Objective:
To develop and internally validate a stacking ensemble ML model for predicting POAF after CABG, identify key predictors, and evaluate its incremental clinical value against conventional models.
Methods:
A retrospective analysis of 563 CABG patients (29.5% POAF) from a single center was performed. After preprocessing, the dataset was split into training (n = 394) and validation (n = 169) sets. Features were selected using elastic net with stability selection (200 bootstrap resamples). Nine base ML algorithms and a stacking ensemble were built; hyperparameters were tuned via fivefold cross‑validation with overfitting controls. Model performance was assessed by discrimination, calibration, and decision curve analysis (DCA).
Results:
Thirteen predictors were retained, with age, intraoperative phenylephrine use, and stroke history ranking highest. The stacking model achieved a validation AUC of 0.7425 and an F1 of 0.711. Logistic regression showed a comparable AUC (0.718, p = 0.09). The three clinical risk scores performed worse (AUCs 0.695, 0.662, 0.670; all p < 0.05). DCA revealed no net benefit below a threshold of 0.4, a marginal benefit (≈0.05) between 0.4 and 0.6, and no benefit above 0.6.
Conclusion:
The stacking model provides only moderate discrimination for POAF after CABG, with negligible net clinical benefit confined to a narrow threshold range. It is not ready for clinical use without external validation. The identified predictors may generate hypotheses for mechanistic studies, but their clinical utility remains unproven.