Implementation of a Heart Disease Risk Prediction Model Using Machine Learning.

Area of Science:

• Medical Informatics
• Computational Biology

Background:

• Cardiovascular disease (CVD) prediction is crucial for timely patient care and lifestyle interventions.
• Machine learning (ML) offers a promising approach to analyze heart disease symptoms and improve diagnostic accuracy.

Purpose of the Study:

• To develop and evaluate machine learning models for predicting cardiovascular disease risk.
• To identify the most effective ML algorithm for heart disease prediction using the Cleveland HD dataset.

Main Methods:

• Feature selection was performed using the chi-square statistical test on the Cleveland heart disease (HD) dataset.
• Several ML algorithms were employed: Support Vector Machine (SVM), Gaussian Naive Bayes, logistic regression, LightGBM, XGBoost, and random forest.
• Model performance was evaluated based on prediction accuracy.

Main Results:

• The random forest algorithm demonstrated the highest accuracy at 88.5% for heart disease risk prediction.
• Other models achieved varying accuracies: SVM (80.32%), Gaussian Naive Bayes (78.68%), logistic regression (80.32%), LightGBM (77.04%), and XGBoost (73.77%).
• Data visualization was used to explore feature relationships.

Conclusions:

• The random forest algorithm is a highly effective tool for cardiovascular disease risk prediction.
• Accurate prediction using ML can significantly aid healthcare practitioners in patient management and early intervention strategies.