Machine Learning for Predicting Coronary Heart Disease Risk in Patients with Hypertension: An Ensemble Modeling...

Area of Science:

• Cardiovascular disease research
• Machine learning in healthcare
• Predictive modeling for chronic conditions

Background:

• Hypertension and coronary heart disease (CHD) are significant global health concerns.
• Early and accurate prediction of CHD in hypertensive patients is crucial for effective management.
• Existing risk stratification methods may benefit from advanced computational approaches.

Purpose of the Study:

• To develop an optimized ensemble learning model for predicting hypertension complicated by CHD.
• To enhance the accuracy and stability of risk assessment for essential hypertension (EH) with CHD.
• To leverage advanced feature selection and classifier fusion techniques.

Main Methods:

• An ensemble model using voting fusion was constructed for early detection of EH complicated by CHD.
• A dataset of 2,487 EH with CHD patients and 3,904 controls was utilized.
• Feature selection identified an 18-dimensional feature set, and five ML algorithms were integrated via voting ensemble.

Main Results:

• The voting fusion ensemble model demonstrated superior performance compared to individual classifiers.
• The model achieved an area under the curve (AUC) of 0.906.
• The prediction accuracy reached 0.888 for EH complicated by CHD.

Conclusions:

• The ensemble model offers improved classification accuracy and robustness for hypertension-associated CHD risk.
• It serves as a clinically useful tool for early risk stratification.
• Further validation is needed, but the framework shows potential as a decision-support tool in clinical informatics.