HEART FAILURE DETECTION USING MACHINE LEARNING ALGORITHMS

Abstract

According to estimates, 17.9 million people succumb to cardiovascular diseases (CVDs) annually, which accounts for 31% of all fatalities globally. The term "cardiovascular diseases" (CVDs) refers to a variety of heart and blood vessel conditions. They mainly constitute heart diseases, coronary artery diseases etc. The terminology "cardiovascular diseases" (CVDs) also refers to a variety of heart and blood vessel conditions. They consist of deep vein thrombosis, pulmonary embolism, rheumatic heart disease, congenital heart disease, peripheral arterial disease, cerebral vascular disease, and coronary heart disease. Cardiovascular disease claims one life in the United States every 34 seconds. Heart disease is the primary cause of mortality in the US for both males and females. Heart disease claimed the lives of almost 697,000 people in 2020, accounting for 1 in 5 fatalities. For the early diagnosis and treatment of those with cardiovascular illness or who are at high cardiovascular risk, a machine learning model can be highly beneficial. (This is primarily caused by the existence of one or more risk factors, such as diabetes, hyperlipidemia, hypertension, or an existing illness). In this work, six classifiers were tried for the dataset and it was concluded that the best models that suited our dataset while modeling was the Naive Bayes Classifier and the Random Forest Classifier with 93.33% Accuracy each.