Effectively Predicting the Presence of Coronary Heart Disease Using Machine Learning Classifiers

Area of Science:

• Cardiology
• Medical Informatics
• Machine Learning in Healthcare

Background:

• Coronary heart disease (CHD) is a leading global cause of mortality.
• Accurate prediction of CHD remains a significant challenge in clinical data analysis.
• Machine learning (ML) offers potential for diagnostic assistance and data-driven prediction in healthcare.

Purpose of the Study:

• To evaluate the effectiveness of eleven ML classifiers for heart disease prediction.
• To identify key features that enhance the predictability of CHD.
• To develop and assess ML models for improved CHD risk assessment.

Main Methods:

• Utilized eleven distinct ML classifiers for heart disease prediction.
• Employed various feature combinations to optimize prediction models.
• Applied well-known classification algorithms to clinical datasets.

Main Results:

• Achieved 95% accuracy using gradient boosted trees and multilayer perceptron models.
• Random Forest classifier demonstrated superior performance with 96% accuracy.
• Identified key features that significantly improved CHD prediction accuracy.

Conclusions:

• ML techniques, particularly Random Forest, are highly effective for CHD prediction.
• Feature selection and optimized ML models can substantially enhance diagnostic accuracy.
• ML-based prediction models show promise for clinical decision support in cardiology.