A robust framework for enhancing cardiovascular disease risk prediction using an optimized category boosting model.

Area of Science:

• Medical Informatics
• Machine Learning
• Cardiology

Background:

• Cardiovascular disease (CVD) is a primary global cause of death.
• Accurate CVD risk prediction is crucial for effective prevention and treatment.
• Machine learning (ML) shows promise in advancing CVD risk assessment.

Purpose of the Study:

• To introduce CVD-OCSCatBoost, a novel ML framework for precise CVD risk prediction.
• To assess various risk factors contributing to cardiovascular disease.
• To enhance the accuracy and efficiency of CVD risk prediction models.

Main Methods:

• Utilized Lasso regression for optimal feature selection.
• Integrated an optimized category-boosting tree (CatBoost) model.
• Developed the opposition-based learning cuckoo search (OCS) algorithm to enhance the CatBoost model, creating OCSCatBoost.

Main Results:

• OCSCatBoost demonstrated superior performance over various ML algorithms.
• Achieved an overall accuracy of 73.67%, recall of 72.17%, and AUC of 0.8024.
• Validated the efficacy of the proposed OCSCatBoost algorithm through extensive comparisons.

Conclusions:

• The CVD-OCSCatBoost framework shows significant potential for improving cardiovascular disease risk prediction.
• This approach can aid in early identification and management of individuals at risk.
• Highlights the advancement of ML applications in cardiovascular health.