Heart Disease Prediction: A Comprehensive Exploration of Optimal Predictive AI

At the forefront of global health challenges, cardiovascular disease is a leading cause of death. Addressing the prevention of this disease effectively is an important effort for human welfare. The diagnostic process usually involves a combination of clinical evaluation, laboratory tests, and various diagnostic methods, making it complex and time-consuming. Conversely, advances in technology and medical research have improved techniques for diagnosing heart disease, potentially improving patient outcomes. Additionally, the application of machine learning (ML) technology has shown promise in improving heart disease diagnosis. This paper proposes a diagnostic support system that relies on optimized machine learning algorithms, which includes various methods, such as Support Vector Machine (SVM), decision tree (DT), K-Nearest Neighbors (K-NN), Naïve Bayes (NB), logistic regression (LR), Random Forest (RF), and Multilayer Perceptron (MLP). The aim of this research is to analyze various components of patient data to make accurate predictions about heart disease. Key factors in diagnosing heart disease include age, gender, smoking habits, obesity, diet, physical activity, stress, type of chest pain, previous chest pain, diastolic blood pressure, diabetes, troponin, and ECG. When compared with other algorithms, multilayer perceptron and logistic regression, which utilize selected features, achieved the highest accuracy rate (89.4%). The proposed method can be a framework for predicting early-stage heart disease.