An Efficient Machine Learning Model Based on Improved Features Selections for Early and Accurate Heart Disease...

Area of Science:

• Cardiology
• Computer Science
• Data Science

Background:

• Coronary heart disease significantly impacts human life, necessitating reliable diagnostic tools.
• Traditional medical history-based diagnosis of heart disease is often unreliable.
• Machine learning (ML) offers a more accurate and efficient approach to cardiac disease detection.

Purpose of the Study:

• To address shortcomings in existing heart disease detection methods.
• To construct an accurate machine learning model for predicting heart disease.
• To evaluate the impact of feature selection on model performance and efficiency.

Main Methods:

• Utilized five feature selection algorithms to optimize a machine learning model.
• Evaluated model performance using metrics like accuracy, precision, recall, F1-score, and MCC.
• Assessed the impact of feature reduction on classifier performance and execution time.

Main Results:

• Support Vector Machine (SVM) achieved 97.5% accuracy.
• K-Nearest Neighbor (KNN) achieved 95% accuracy.
• Logistic Regression achieved 93% accuracy, with reduced computation times for all models.

Conclusions:

• Feature reduction positively impacts classifier performance and reduces computation time.
• The proposed ML model demonstrates high accuracy and efficiency in heart disease prediction.
• Accurate and precise heart disease detection using ML can help prevent human loss.