Heart disease kills the most Americans every year and Coronary Artery Disease (CAD) is the most common form of heart disease. Early detection and prevention of CAD can significantly reduce mortality but currently, CAD is diagnosed using expensive and invasive imaging tests like coronary angiography. In this study, we used a two-staged machine learning-based prediction model to predict the status of CAD using minimal imaging data. Data mining is the process of defining a pattern from the given set of data. This project applied pattern-finding techniques to predict the status of CAD in human patients. We successfully predicted blockage of specific coronary arteries using the first stage of our model. Our models could serve as a blueprint for developing a more robust predictive device or software that provides an inexpensive, early indicator for this life-threatening disease.