Heart Disease Detection Using Machine Learning Techniques

Technology is growing rapidly and entering into almost all fields including health care. Coronary heart diseases (CHDs) are accounting for the larger number of deaths, which are estimated to be 17 million, i.e., 34% of the global death. Heart disease is considered to be one of the riskiest diseases, yet considered the most manageable disease. Hence, if the disease could be predicted early, it could be more beneficial. Here, ‘Artificial Intelligence (AI) and Machine Learning (ML)’ come into action as data science contributes a significant role in the field of healthcare, as it helps in analyzing and processing huge data. This paper explains how based on medical data and habits of a person, an ML model can predict whether a person will be a risk of CHD in the next 10 years or not. This experiment tends to find out the most accurate model for heart disease prediction by using various classification techniques of ML such as artificial neural networks (ANNs), logistic regression (LR), random forest (RF), and decision tree (DT). It is thus found that RF has the highest accuracy.