Risk Prediction Of Coronary Heart Disease Using A Decision Tree Algorithm Based On Patient Medical Records

Abstract

Coronary heart disease (CHD) remains one of the leading causes of death worldwide, often due to late diagnosis and inadequate early detection. Early risk prediction of CHD is crucial to improve patient outcomes and reduce mortality. This study aims to develop a predictive model for assessing the risk of coronary heart disease using a decision tree algorithm, based on structured patient medical records. The dataset used contains various clinical features, including age, gender, cholesterol level, blood pressure, blood sugar, ECG results, and exercise-induced angina. A decision tree classifier was selected for its interpretability, ease of implementation, and effectiveness in handling categorical and numerical data. Data preprocessing steps such as missing value handling, normalization, and feature selection were applied to improve model performance. The model was trained and validated using k-fold cross-validation to ensure reliability. Performance was evaluated based on accuracy, precision, recall, and F1-score. The results demonstrate that the decision tree algorithm achieved satisfactory performance in predicting CHD risk, making it a potentially valuable tool for supporting clinical decision-making. This study highlights the importance of integrating data mining techniques into healthcare to enable timely and accurate risk assessment of life-threatening diseases such as coronary heart disease.