Mortality Risk of People With/Without Heart Conditions Using Machine Learning Heart disease is a very large problem in the United States and around the world. People can get their regularly scheduled doctor visits every and still have no idea what their potential risk for heart disease until it’s too late. The fact that this is the leading cause of death in the US illustrates that the benefit of an algorithm that could predict someone’s risk of death by quickly analyzing easily obtainable metrics would be astronomical. The goal here is to create a Machine Learning model that can accomplish our task while also using an imbalanced dataset. The methods used in the project were a simple regression, a boosted trees model, and a keras sequential model. Current research has shown great promise in this direction with researchers being able to achieve accuracy score in the high 90s. We want to see if we can do the same using similar but limited to achieve the same results as the researchers. Also as healthcare costs continue to increase being able to provide this kind of information to doctors and patients alike would allow to make those costs more manageable.