Explainable machine learning model for predicting the occurrence of postoperative malnutrition in children with
Hui Shi1, Dong Yang2, Kaichen Tang2
1Guangzhou Women and Children's Medical Center, Institute of Pediatrics, Guangzhou Medical University, No.9 Jinsui Road, Zhujiang Newtown, Tianhe District, Guangzhou, 510623, China; Department of Biostatistics and Epidemiology, School of Public Health, Sun Yat-sen University, Guangzhou, China.
Insights
Machine learning models accurately predict malnutrition in children with congenital heart disease (CHD) one year after surgery. Explainable AI helps clinicians understand predictions for tailored nutritional interventions.
Area of Science:
- Pediatric Cardiology
- Computational Biology
- Nutritional Science
Background:
- Malnutrition affects 50-75% of children post-congenital heart disease (CHD) surgery, necessitating early prediction for intervention.
- Predicting malnutrition is crucial for timely nutritional support in pediatric patients with CHD.
Purpose of the Study:
- To develop and validate machine learning (ML) models for predicting malnutrition in children with CHD.
- To utilize explainable ML methods to interpret model predictions and identify key risk factors for malnutrition.
Main Methods:
- A prospective cohort study of 536 children with CHD undergoing complete repair from December 2017 to May 2020.
- Development of five ML prediction models using 115 features, with performance evaluated by AUC.
- Application of permutation importance and SHapley Additive exPlanations (SHAP) for feature interpretation.
Main Results:
- The XGBoost ML model demonstrated the highest AUC for predicting underweight, stunted, and wasting statuses one year post-surgery.
- Key predictors for underweight status included WAZ-scores (preoperative, 1-month, and discharge).
- Predictors for stunted status included HAZ scores and aortic clamping time; wasting status predictors included hospital length of stay and formula intake.
Conclusions:
- An explainable XGBoost ML model accurately predicts long-term malnutrition in pediatric CHD patients post-surgery.
- The model's explainability aids clinicians in understanding malnutrition risk factors.
- Findings support individualized nutritional follow-up strategies for children with CHD.
Background & Aims:
Malnutrition is persistent in 50%-75% of children with congenital heart disease (CHD) after surgery, and early prediction is crucial for nutritional intervention. The aim of this study was to develop and validate machine learning (ML) models to predict the malnutrition status of children with CHD. We used explainable ML methods to provide insight into the model's predictions and outcomes.
Methods:
This prospective cohort study included consecutive children with CHD admitted to the hospital from December 2017 to May 2020. The cohort data were divided into the training and test data sets based on the follow-up time. The outcome of the study was CHD child malnutrition 1 year after surgery, the primary outcome was an underweight status, and the secondary outcomes were stunted and wasting status. We used five ML algorithms with multiple features to construct prediction models, and the performance of these ML models was measured by an area under the receiver operating characteristic curve (AUC) analysis. We also used the permutation importance and SHapley Additive exPlanations (SHAP) to determine the importance of the selected features and interpret the ML models.
Results:
We enrolled 536 children with CHD who underwent complete repair. The proportions of children with an underweight, stunted, or wasting status 1 year after surgery were 18.1% (97/536), 12.1% (65/536), and 17.5% (94/536), respectively. All patients contributed to the generation of 115 useable features, which allowed us to build models to predict malnutrition. Five prediction algorithms were used, and the XGBoost model achieved the greatest AUC in all outcomes. The results obtained from the permutation importance and SHAP analyses showed that the 1-month postoperative WAZ-score, discharge WAZ score and preoperative WAZ score were the top 3 important features in predicting an underweight status in the XGBoost algorithm. Regarding the stunted status, the top 3 important features were the 1-month postoperative HAZ score, discharge HAZ score, and aortic clamping time. Regarding the wasting status, the top 3 important features were the hospital length of stay, formula intake, and discharge WHZ-score. We also used a narrative case report as an example to describe the clinical manifestations and predicted the primary outcomes of two children.
Conclusions:
We developed an ML model (XGBoost) that provides accurate early predictions of malnutrition 1-year postoperatively in children with CHD. Because the ML model is explainable, it may better enable clinicians to better understand the reasoning underlying the outcome. Our study could aid in determining individual treatment and nutritional follow-up strategies for children with CHD.
