Enhanced Evolutionary Feature Selection and Ensemble Method for Cardiovascular Disease Prediction.

Area of Science:

• Medical Informatics
• Machine Learning
• Cardiology

Background:

• Cardiovascular Disease (CVD) is a leading global cause of mortality.
• Accurate CVD prediction remains a challenge in medical data analysis.
• Existing models often lack focus on critical feature selection for CVD prediction.

Purpose of the Study:

• To identify key features and machine learning methods for enhanced CVD prediction.
• To develop a reliable and efficient model for diagnosing heart disease.
• To improve the precision, recall, and accuracy of cardiovascular disease classification.

Main Methods:

• Utilized Big Data and Artificial Intelligence for automated model development.
• Implemented an enhanced evolutionary feature selection technique.
• Developed a hybrid ensemble model for classification and diagnosis.

Main Results:

• The proposed model achieved high classification accuracy: 93.65% (statlog), 82.81% (SPECTF), and 84.95% (coronary heart disease).
• The enhanced feature selection and hybrid ensemble model outperformed existing methods.
• Performance was validated against state-of-the-art machine learning techniques using ROC curves.

Conclusions:

• The developed approach significantly enhances the prediction capability for cardiovascular disease.
• Feature selection is crucial for improving the accuracy of heart disease diagnosis models.
• This work offers a more reliable and efficient method for CVD prediction.