A comparative study of ensemble learning based classifiers in heart disease detection

Area of Science:

• Cardiology
• Machine Learning
• Data Science

Background:

• Heart disease remains a leading global health challenge.
• Accurate and early detection are crucial for effective management and improved patient survival rates.
• Ensemble learning methods offer potential for enhancing classification accuracy in medical diagnostics.

Purpose of the Study:

• To evaluate and compare the performance of various ensemble learning algorithms for heart disease classification.
• To identify the most effective algorithms for improving the accuracy of early heart disease detection.
• To contribute to the development of more reliable diagnostic tools for cardiovascular conditions.

Main Methods:

• A comparative analysis of multiple ensemble learning algorithms including Gradient Boosting, Random Forest, AdaBoost, XGBoost, Bagging, Extra Trees, Voting, Stacking, HistGradientBoosting, and LightGBM.
• Utilizing comprehensive evaluation metrics such as confusion matrix, precision, recall, and F1-score for performance assessment.
• Applying these methods to a dataset for heart disease classification.

Main Results:

• Performance metrics indicated significant differences in classification accuracy among the evaluated ensemble methods.
• Specific algorithms demonstrated superior performance in identifying heart disease cases.
• The study provides insights into the efficacy of different ensemble techniques for this specific medical application.

Conclusions:

• Ensemble learning algorithms show significant promise for enhancing heart disease classification accuracy.
• The findings can guide the selection of optimal models for developing advanced, precise early detection systems.
• Further research can build upon these results to refine diagnostic tools and improve cardiovascular care.