.png){kind=icon}
.png){kind=icon}
Return
Development of an Explainable Machine Learning Model for Cardiovascular-Kidney-Metabolic Syndrome Prediction Based on Dietary Antioxidants in a National Population
.png){kind=icon}
OA
AI
M
H
M
X
W
X
C
DOI：10.1159/000549326.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Introduction: The role of dietary antioxidants in preventing or delaying the progression of cardiovascular-kidney-metabolic (CKM) syndrome remains underexplored. We aimed to develop and interpret a machine learning (ML) model to predict advanced CKM stages based on dietary antioxidant profiles. Methods: Data were analyzed from 10,257 adults aged >30 years in the NHANES 2007-2010 and 2017-2018 cycles. Dietary antioxidant intake was estimated using two 24-h dietary recalls. Five ML algorithms were trained with rigorous hyperparameter optimization and evaluated comprehensively. SHapley Additive exPlanations (SHAP) was applied to elucidate feature importance and individual-level contributions. An online prediction tool was deployed to enhance clinical utility. Results: The eXtreme Gradient Boosting (XGBoost) model achieved the highest predictive performance, yielding an area under the curve of 0.901. SHAP analysis identified seven key predictors: age, sex, smoking status, magnesium, zinc, myricetin, and catechin. Older age, male sex, and smoking were associated with increased CKM risk, whereas higher intakes of magnesium, myricetin, zinc, and catechin were protective. Conclusions: XGBoost effectively predicted advanced CKM stages using a concise set of seven features. Explainable AI approaches such as SHAP enhance model transparency and clinical translation, supporting personalized CKM risk stratification based on dietary antioxidant patterns.
Keywords:
Cardiovascular-kidney-metabolic syndrome
Dietary antioxidants
Machine learning
SHapley Additive exPlanations
eXtreme Gradient Boosting
Journal
J
IF:
2.3
Papers:
17
Citations:
0
Organization
Cited Papers