Building and validating a machine learning model to predict coronary heart disease risk based on non-invasive...

Area of Science:

• Cardiology
• Machine Learning
• Public Health

Background:

• Coronary heart disease (CHD) is a major global cause of mortality.
• Early identification and intervention are critical for managing CHD.
• This study focuses on developing a machine learning (ML) model for CHD risk prediction.

Purpose of the Study:

• To develop and evaluate a predictive model for coronary heart disease (CHD) risk.
• To identify key clinical features contributing to CHD risk.
• To create a user-friendly tool for early CHD screening.

Main Methods:

• Utilized the Behavioral Risk Factor Surveillance System (BRFSS) dataset for model development and internal validation.
• Employed eight machine learning algorithms, including Light Gradient Boosting Machine (LightGBM).
• Validated the optimal model externally using the National Health and Nutrition Examination Survey (NHANES) dataset and SHAP analysis for feature importance.

Main Results:

• The LightGBM model achieved an AUC of 0.825 in internal validation and 0.851 in external validation.
• Key predictors identified include age, sex, hypertension, and dyslipidemia.
• A web-based calculator was developed for CHD risk prediction based on the LightGBM model.

Conclusions:

• The LightGBM-based CHD risk prediction model demonstrates high accuracy.
• This model shows significant potential for early screening and prevention of coronary heart disease.
• The developed tool can assist healthcare professionals in identifying high-risk individuals.