Machine learning techniques for heart disease prediction

Abstract

According to WHO (World Health Organization), Heart diseases are the reason for 12 million deaths every year. In most of the countries, half of the deaths are due to cardiovascular diseases. The early diagnosis of cardiovascular sicknesses can help in settling on choices on the way of life changes in high hazard patients and thusly diminish the difficulties. In this paper, machine learning techniques are used for the detection of heart disease. We also applied sampling techniques for handling unbalanced datasets. Various machine learning methods are used to predict the overall risk. The framingham_heart_disease dataset is publically available on the Kaggle. This dataset is used in our experiments. The end goal is to predict whether the patient has a 10-year risk of future coronary heart disease (CHD). The dataset contains 15 features that give patient information. By applying machine learning techniques, we achieved 99% accuracy in heart disease detection.