IMR Press / RCM / Volume 23 / Issue 11 / DOI: 10.31083/j.rcm2311376
Open Access Original Research
Machine Learning Model for Predicting Risk of In-Hospital Mortality after Surgery in Congenital Heart Disease Patients
Show Less
1 Department of Cardiothoracic Surgery, Shanghai Children’s Medical Center, School of Medicine, Shanghai Jiaotong University, 200127 Shanghai, China
2 Information Center, Shanghai Children’s Medical Center, School of Medicine, Shanghai Jiaotong University, 200127 Shanghai, China
3 Shanghai Synyi Medical Technology Co., Ltd. 201203 Shanghai, China
*Correspondence: (Zhaohui Lu)
These authors contributed equally.
Academic Editor: Jerome L. Fleg
Rev. Cardiovasc. Med. 2022, 23(11), 376;
Submitted: 24 May 2022 | Revised: 27 July 2022 | Accepted: 26 August 2022 | Published: 3 November 2022
Copyright: © 2022 The Author(s). Published by IMR Press.
This is an open access article under the CC BY 4.0 license.

Background: A machine learning model was developed to estimate the in-hospital mortality risk after congenital heart disease (CHD) surgery in pediatric patient. Methods: Patients with CHD who underwent surgery were included in the study. A Extreme Gradient Boosting (XGBoost) model was constructed based onsurgical risk stratification and preoperative variables to predict the risk of in-hospital mortality. We compared the predictive value of the XGBoost model with Risk Adjustment in Congenital Heart Surgery-1 (RACHS-1) and Society of Thoracic Surgery-European Association for Cardiothoracic Surgery (STS-EACTS) categories. Results: A total of 24,685 patients underwent CHD surgery and 595 (2.4%) died in hospital. The area under curve (AUC) of the STS-EACTS and RACHS-1 risk stratification scores were 0.748 [95% Confidence Interval (CI): 0.707–0.789, p < 0.001] and 0.677 (95% CI: 0.627–0.728, p < 0.001), respectively. Our XGBoost model yielded the best AUC (0.887, 95% CI: 0.866–0.907, p < 0.001), and sensitivity and specificity were 0.785 and 0.824, respectively. The top 10 variables that contribute most to the predictive performance of the machine learning model were saturation of pulse oxygen categories, risk categories, age, preoperative mechanical ventilation, atrial shunt, pulmonary insufficiency, ventricular shunt, left atrial dimension, a history of cardiac surgery, numbers of defects. Conclusions: The XGBoost model was more accurate than RACHS-1 and STS-EACTS in predicting in-hospital mortality after CHD surgery in China.

congenital heart disease
in-hospital mortality
machine learning
Extreme Gradient Boosting
Fig. 1.
Back to top