Machine Learning-based Application for Predicting Hearth Disease Risk using CreateML

Heart disease is the leading cause of death in Indonesia. Although predominantly affecting the elderly, it also significantly impacts young and productive individuals. The lack of knowledge about the risks and causes of heart disease contributes to this issue. Furthermore, the high cost of health tests to detect the risk of heart disease increases the problem. Afford-able and effective technology is needed to detect heart disease early. This research aims to develop artificial intelligence-based software for detecting heart disease in productive age groups. The method utilizes machine learning algorithms, particularly for classification problems using CreateML. This work can predict heart disease risk using data from more than 300.000 samples from the Centers for Disease Control and Prevention. Using a random forest algorithm, this research achieved 98.2 % training and 83% testing accuracy. The generated model was deployed as a mobile application to ensure its usability.