Early Prediction of Chronic Heart Disease Based on Electronic Triage Dataset by using Machine Learning

Chronic Heart Disease (CHD) is the main reason for mortality around the world, constituting about 15% of all natural deaths, which confirms the importance of early detection to mitigate these rates. CHD can be detected by measuring multiple health parameters such as blood pressure, SpO2, ECG, and other parameters. Therefore, there is a persistent need for effective and precise prediction methods. Traditional techniques for prediction and analysis are limited due to the complexity of the data and correlations. First approach for developing these techniques is by using machine learning (ML) algorithms. ML can examine vast amounts of patient information, including medical records, test outcomes, and imaging scans to identify patterns and predict the possibility of developing heart disease or having a cardiac event. This study aims to improve the accuracy of prediction by utilizing ML techniques to predict CHD using medical data, reduce the incidence of misdiagnosis, and ultimately save lives. Supervised learning techniques, such as Support Vector Machine (SVM), Artificial Neural Networks (ANN), AdaBoost, Logistic Regression (LR), K-Nearest Neighbor (KNN) and others are employed to identify correlations in CHD data and enhance prediction rates. The results indicate that SVM-derived probabilistic models outperform other algorithms in predicting CHD and the accuracy of the model was 96.2%, as revealed through various performance evaluation metrics.