Predictive Analytics in Cardiology: Machine Learning Techniques for Heart Disease

According to recent studies by the WHO, there is a growing prevalence of conditions associated with coronary heart disease, resulting in approximately 17.9 million deaths annually. With an expanding population, early identification and treatment of such issues become increasingly challenging. Nevertheless, advancements in machine learning techniques have shown promising results in addressing these challenges. Hence, the focus of this paper is to develop a machine-learning model for predicting heart disease based on relevant features. To achieve this, the UCI Heart Disease Prediction dataset is utilized, comprising 14 parameters related to heart health. Various machine learning algorithms including Random Forest, Support Vector Machine (SVM), Naive Bayes, and Decision Tree are employed in the model's development. Additionally, we explore correlations among the dataset features using established machine-learning techniques to enhance predictive accuracy. The results indicate that Random Forest outperforms other algorithms in terms of both accuracy and computational efficiency. As a decision support tool, this model holds promise for aiding clinical practitioners in their facilities.