JMIR Publications

ABSTRACT

Background:

Hyperhomocysteinemia (HHcy) is recognized as an independent risk factor for coronary heart disease (CHD), yet accurately predicting CHD risk in HHcy patients remains a challenge. This study aimed to develop and validate multiple machine learning models for predicting CHD risk in HHcy patients and elucidate key predictors using SHAP algorithms.

Objective:

To develop and validate machine learning models for predicting the risk of coronary heart disease in individuals with normal homocysteine levels, aiming to improve early risk stratification and clinical decision-making.

Methods:

This single-center retrospective study collected data from HHcy-diagnosed patients through electronic medical records, which were randomly divided into training (70%), validation (15%), and test (15%) sets. Seven machine learning models were constructed, including Logistic Regression, KNN, Decision Tree, Random Forest, Extreme Gradient Boost, LightGBM, and Stacking. Six core variables (age, weight, hypertension, continuous drinking history, APTT, and carotid plaque) were utilized as inputs, with performance evaluation metrics encompassing AUC, accuracy, F1 score, calibration curve, Brier score, and decision curve analysis. Additionally, SHAP interpretation was conducted on the optimal LightGBM model.

Results:

The LightGBM model exhibited superior performance in the test set (AUC = 0.807, F1 score = 0.606), demonstrated good calibration (Brier score = 0.2415), and yielded high clinical net benefit. SHAP analysis revealed age and APTT as the most influential predictors, followed by hypertension, weight, carotid plaque, and continuous drinking history. The correlation heat map illustrated low collinearity among variables, ensuring model stability.

Conclusions:

The LightGBM model demonstrated high accuracy and interpretability in forecasting CHD risk among HHcy patients. The integration of machine learning and interpretable artificial intelligence methods holds promise for delivering personalized early risk assessment and intervention strategies in clinical settings Clinical Trial: This study is a retrospective modeling analysis and does not involve a registered clinical trial. Therefore, trial registration is not applicable.