Efficient Data-Driven Machine Learning Models for Cardiovascular Diseases Risk Prediction

Area of Science:

• Cardiology
• Medical Informatics
• Machine Learning

Background:

• Cardiovascular diseases (CVDs) are the leading cause of death globally.
• CVDs encompass conditions like hypertension, heart failure, myocardial infarction, and stroke.
• Early diagnosis and prevention are crucial for managing CVDs and improving patient outcomes.

Purpose of the Study:

• To develop efficient machine learning (ML) models for predicting cardiovascular disease (CVD) manifestation.
• To evaluate the effectiveness of the Synthetic Minority Oversampling Technique (SMOTE) in improving CVD prediction models.
• To identify key risk factors contributing to CVD prediction.

Main Methods:

• A supervised ML methodology was employed for CVD prediction.
• Various ML models were trained and tested using identified risk factors.
• The Synthetic Minority Oversampling Technique (SMOTE) was applied to address class imbalance, and model performance was compared using Accuracy, Recall, Precision, and AUC.

Main Results:

• The Stacking ensemble model, combined with SMOTE and 10-fold cross-validation, achieved the highest performance.
• This model reported an Accuracy of 87.8%, Recall of 88.3%, Precision of 88%, and an Area Under the Curve (AUC) of 98.2%.
• The results highlight the superiority of SMOTE in enhancing ML model performance for CVD prediction.

Conclusions:

• The study demonstrates the potential of ML, particularly the Stacking ensemble model with SMOTE, for accurate CVD prediction.
• Effective risk factor identification and utilization are vital for robust predictive models.
• The findings suggest that this approach can significantly aid in the early diagnosis and prevention of cardiovascular diseases.