Predictive Value of Obesity Indices for Carotid Atherosclerosis: An Interpretable Machine Learning Study
1
College of Public Health, Zhengzhou University, Zhengzhou, China
Popul. Med. 2026;8(Supplement Supplement 1):
ABSTRACT
INTRODUCTION:
Obesity is a key risk factor for cardiovascular diseases but the efficacy of different obesity indices varies. This study aimed to evaluate the comparative predictive value of six commonly used obesity indices for carotid atherosclerosis (CAS) using interpretable machine learning algorithms and to identify an optimal low-cost screening tool for high-risk populations.
METHODS:
A large-scale cross-sectional study was conducted between August 2021 and August 2022 involving 36376 adults across Henan Province China. Participants underwent standardized physical examinations biochemical assays and carotid ultrasonography. Six indices including body mass index (BMI) waist circumference (WC) waist-to-hip ratio (WHR) waist-to-height ratio (WHtR) a body shape index (ABSI) and body roundness index (BRI) were analyzed. The Boruta algorithm was used for feature selection. Non-linear relationships were modeled using restricted cubic spline analysis. Four machine learning models including SVM RF XGBoost and LightGBM were trained and validated via ten-fold cross-validation. Model interpretability was achieved using SHapley Additive exPlanations (SHAP).
RESULTS:
The prevalence of CAS was 49.6%. Among the models XGBoost demonstrated superior predictive performance. While all indices were significantly associated with CAS (P<0.001) WHtR emerged as the most effective predictor. The combination of WHtR with the optimal feature subset in the XGBoost model achieved the highest area under the curve of 0.8824. SHAP analysis revealed that age gender hypertension history obesity indices and physical activity were the top contributors to CAS risk.
CONCLUSIONS:
Advanced machine learning models significantly enhance CAS risk prediction. Among the anthropometric indicators WHtR outperformed traditional BMI. Given its high predictive accuracy non-invasiveness and cost-effectiveness WHtR is recommended as a priority screening tool for carotid atherosclerosis in routine primary care and large-scale public health screenings particularly in resource-limited settings.
Obesity is a key risk factor for cardiovascular diseases but the efficacy of different obesity indices varies. This study aimed to evaluate the comparative predictive value of six commonly used obesity indices for carotid atherosclerosis (CAS) using interpretable machine learning algorithms and to identify an optimal low-cost screening tool for high-risk populations.
METHODS:
A large-scale cross-sectional study was conducted between August 2021 and August 2022 involving 36376 adults across Henan Province China. Participants underwent standardized physical examinations biochemical assays and carotid ultrasonography. Six indices including body mass index (BMI) waist circumference (WC) waist-to-hip ratio (WHR) waist-to-height ratio (WHtR) a body shape index (ABSI) and body roundness index (BRI) were analyzed. The Boruta algorithm was used for feature selection. Non-linear relationships were modeled using restricted cubic spline analysis. Four machine learning models including SVM RF XGBoost and LightGBM were trained and validated via ten-fold cross-validation. Model interpretability was achieved using SHapley Additive exPlanations (SHAP).
RESULTS:
The prevalence of CAS was 49.6%. Among the models XGBoost demonstrated superior predictive performance. While all indices were significantly associated with CAS (P<0.001) WHtR emerged as the most effective predictor. The combination of WHtR with the optimal feature subset in the XGBoost model achieved the highest area under the curve of 0.8824. SHAP analysis revealed that age gender hypertension history obesity indices and physical activity were the top contributors to CAS risk.
CONCLUSIONS:
Advanced machine learning models significantly enhance CAS risk prediction. Among the anthropometric indicators WHtR outperformed traditional BMI. Given its high predictive accuracy non-invasiveness and cost-effectiveness WHtR is recommended as a priority screening tool for carotid atherosclerosis in routine primary care and large-scale public health screenings particularly in resource-limited settings.
| eISSN: | 2654-1459 |
We process personal data collected when visiting the website. The function of obtaining information about users and their behavior is carried out by voluntarily entered information in forms and saving cookies in end devices. Data, including cookies, are used to provide services, improve the user experience and to analyze the traffic in accordance with the Privacy policy. Data are also collected and processed by Google Analytics tool (more).
You can change cookies settings in your browser. Restricted use of cookies in the browser configuration may affect some functionalities of the website.
You can change cookies settings in your browser. Restricted use of cookies in the browser configuration may affect some functionalities of the website.
