Computer-Aided Diagnostics of Heart Disease Risk Prediction Using Boosting Support Vector Machine

Area of Science:

• Cardiology
• Machine Learning
• Medical Informatics

Background:

• Heart disease is a major global cause of mortality, necessitating advancements in early detection.
• Accurate risk stratification is vital for timely intervention and improved patient prognosis.

Purpose of the Study:

• To develop and evaluate a boosting Support Vector Machine (SVM) model for enhanced heart disease risk prediction.
• To compare the performance of the proposed model against other established machine learning algorithms.

Main Methods:

• Utilized a Cleveland clinic dataset with 13 attributes and 303 records, with missing values handled via listwise deletion.
• Employed feature selection using a boosting technique to enhance model efficiency and accuracy.
• Implemented a train/test split for data partitioning, followed by SVM model training and evaluation with a linear kernel and C=0.05.

Main Results:

• The boosting SVM model demonstrated superior performance compared to Logistic Regression, Nave Bayes, Decision Trees, Multilayer Perceptron, and Random Forest.
• Feature selection via boosting improved model accuracy and reduced computational time.

Conclusions:

• The proposed boosting SVM approach offers a more accurate and efficient tool for computer-aided diagnosis of heart disease risk.
• This method holds significant potential for improving early detection and management strategies for cardiovascular conditions.