SHAP-LR: An Interpretable Logistic Regression Model for Coronary Heart Disease Risk Prediction

Introduction : Coronary Heart Disease (CHD) is a global leading cause of death, demanding accurate predictive tools for early intervention. This study develops an interpretable machine learning framework (SHAP-LR) for CHD risk prediction, combining feature engineering, logistic regression, and SHAP-based explainability to support clinical decision-making. Methods : The study employed three publicly accessible datasets: BRFSS_2015, Cleveland, Hungary, Switzerland, VA Long Beach, and Stalog (Heart) datasets. Data preprocessing involved cleaning, standardization, and feature selection, with SHAP values used to enhance interpretability. Multiple machine learning models, including Decision Tree, AdaBoost, Gradient Boost, Bagging, CatBoost, Extra Trees, and Logistic Regression, were evaluated. Model performance was assessed using accuracy, precision, recall, F1-score, and AUROC. A user-friendly HD scoring system was developed based on the best-performing model, logistic regression, which was further optimized through hyperparameter tuning. Results : Logistic regression outperformed other models, achieving an accuracy of 90.54% and an AUROC of 80.27% on the BRFSS_2015 dataset. After hyperparameter tuning, the model's performance improved further, with accuracy reaching 90.55% and AUROC increasing to 81.09%. The SHAP value analysis revealed that age, high blood pressure, and high cholesterol were the most significant predictors of HD. The developed scoring system provided a quantifiable risk assessment tool, enabling clinicians to stratify patients based on their HD risk effectively. Conclusion : This study demonstrates the effectiveness of machine learning in predicting heart disease, with logistic regression emerging as the most reliable model. The integration of SHAP values enhanced the model's interpretability, making it a valuable tool for clinical decision-making. The developed HD scoring system offers a practical and efficient method for risk assessment, potentially improving early diagnosis and intervention strategies. Future work will focus on validating the model in diverse clinical settings and exploring additional features to further enhance predictive accuracy.