Machine Learning Models for Cardiovascular Disease Events Prediction.

Area of Science:

• Cardiovascular medicine
• Biostatistics
• Machine learning in healthcare

Background:

• Cardiovascular diseases (CVDs) represent a significant global health burden, contributing to high mortality rates worldwide.
• Early diagnosis and prevention strategies are crucial for managing CVDs, necessitating the identification of reliable risk biomarkers.
• Machine learning (ML) offers powerful tools for analyzing complex clinical and biochemical data to predict disease outcomes.

Purpose of the Study:

• To evaluate the efficacy of various machine learning models in predicting 10-year cardiovascular disease (CVD) mortality.
• To identify the most accurate ML algorithm for CVD risk assessment using clinical and biochemical data.
• To provide a foundation for risk stratification tools like the TIMELY study.

Main Methods:

• Utilized the Ludwigshafen Risk and Cardiovascular Health (LURIC) study cohort, including 2943 patients (484 deceased due to CVD).
• Applied multiple machine learning models: Logistic Regression (LR), Support Vector Machine (SVM), Random Forest (RF), Naïve Bayes (NB), Extreme Grading Boosting (XGB), and Adaptive Boosting (AdaBoost).
• Assessed model performance using metrics such as Accuracy (ACC), Precision, Recall, F1-Score, Specificity (SPE), and Area Under the ROC Curve (AUC).

Main Results:

• Logistic Regression (LR) demonstrated the highest predictive performance with an accuracy of 72.20%.
• Comparative analysis evaluated the performance of six distinct machine learning algorithms for CVD mortality prediction.
• The study identified specific ML models suitable for clinical application in risk assessment.

Conclusions:

• Logistic Regression is a reliable algorithm for predicting 10-year CVD mortality based on clinical and biochemical data.
• The findings support the integration of ML-driven risk scores for proactive CVD management.
• These results will inform the TIMELY study for estimating CVD risk and mortality in patient populations.