Development and Validation of a Predictive Model for Coronary Artery Disease Using Machine Learning.

Area of Science:

• Cardiology
• Artificial Intelligence
• Medical Informatics

Background:

• Early identification of coronary artery disease (CAD) is crucial for preventing disease progression and reducing mortality.
• Conventional risk factors and laboratory test data are key indicators for CAD risk assessment.

Purpose of the Study:

• To construct and validate a machine learning model for predicting CAD risk.
• To leverage Random Forest algorithm for enhanced predictive accuracy.

Main Methods:

• Utilized data from 3,112 CAD patients and 3,182 controls across three Chinese centers.
• Employed the Random Forest algorithm to build a predictive model for CAD.
• Assessed model performance using receiver operating characteristic (ROC) curves and key metrics like AUC, sensitivity, and specificity.

Main Results:

• The Random Forest model demonstrated high predictive capability in the development cohort (AUC 0.948).
• Validation cohorts confirmed the model's strong discriminatory ability (AUCs 0.944 and 0.940).
• The model achieved high sensitivity and specificity across all cohorts, indicating reliable CAD risk prediction.

Conclusions:

• A validated Random Forest model effectively predicts CAD risk using 15 clinical and lab indexes.
• The developed tool offers a valuable resource for clinical practice in CAD management and primary prevention.
• Early and accurate CAD risk assessment facilitates timely intervention and improves patient outcomes.