Development and validation of an explainable machine learning-based prediction model for primary Kawasaki disease
Zixia Song1,2, Hongjun Ming1, Bin Liu2
1Department of Pediatrics, Beijing Anzhen Nanchong Hospital, Capital Medical University (Nanchong Central Hospital), Nanchong, China.
Insights
An explainable machine learning model using XGBoost can predict coronary artery aneurysms (CAA) in children with Kawasaki disease (KD). This tool aids in early risk identification for better patient outcomes.
Area of Science:
- Pediatric Cardiology
- Machine Learning in Medicine
- Medical Informatics
Background:
- Kawasaki disease (KD) can lead to coronary artery aneurysms (CAA) in up to 20% of untreated children.
- Intravenous immunoglobulin (IVIG) therapy is standard for acute KD but does not eliminate CAA risk.
Purpose of the Study:
- To develop and validate an explainable machine learning (ML) model for predicting CAA in KD patients.
- To identify key predictors of CAA development in KD.
Main Methods:
- Retrospective analysis of clinical data from 327 KD patients (2015-2023).
- Development and comparison of six ML models: SVM, KNN, Lasso, XGBoost, RF, MLP.
- Validation of the best-performing model using SHapley Additive exPlanations (SHAP) for interpretability.
Main Results:
- The XGBoost-based ML model demonstrated optimal performance in predicting CAA.
- Receiver operator characteristic (ROC) curve analysis confirmed XGBoost's superior predictive accuracy.
- SHAP analysis provided insights into individual sample predictions based on variable contributions.
Conclusions:
- An explainable XGBoost model shows promise for predicting CAA in children with KD.
- This ML approach can aid clinicians in identifying high-risk patients for targeted interventions.
Background:
Kawasaki disease (KD) can lead to coronary artery aneurysms (CAA) in approximately 1 in 5 untreated children despite intravenous immunoglobulin (IVIG) therapy in the acute phase. The aim of this study is to develop and validate an explainable machine learning (ML)-based prediction model for CAA in KD.
Methods:
This study retrospectively analyzed the clinical data of children diagnosed with primary KD at Nanchong Central Hospital, Sichuan Province between 2015 and 2023. Six models, including support vector machine (SVM), K-nearest neighbors (KNN), least absolute shrinkage and selection operator (Lasso), extreme gradient boosting (XGBoost), random forest (RF), and multilayer perceptron (MLP), based on ML algorithms were developed. The model with optimal performance was validated and the explainable SHapley Additive exPlanations (SHAP) analysis was used.
Results:
A total of 327 children diagnosed with KD were included in the training set and validation set. Receiver operator characteristic curve analysis showed that XGBoost based model exhibited an optimal performance among the six models. Moreover, for a given CAA positive sample, the sum of the SHAP values of all variables of XGBoost represented the individual deviation from the mean predicted from the entire dataset.
Conclusions:
The XGBoost algorithm-based explainable model might be used to predict the occurrence of CAA in children with KD.
More Related Videos
06:16Signal Acquisition, Score Interpretation, and Economics of a Non-Invasive Point-of-Care Test for Coronary Artery Disease
Published on: August 9, 2024
07:35A Knowledge Graph Approach to Elucidate the Role of Organellar Pathways in Disease via Biomedical Reports
Published on: October 13, 2023