Prediction of heart disease using long short-term memory based network

Abstract

Technology has become one of the greatest needs of humankind and, in every field, has made lives better. In the field of medicine, technology has helped to treat, diagnose, and cure the diseases far better than the traditional method. Atherosclerosis is a condition where the cholesterol in the body gets deposited on the walls of blood vessels, thus narrowing them and reducing the blood flow to the organs. If the rate of atherosclerosis is greater than 50%, the patient stands at a high risk of heart diseases. The healthcare industry collects a large amount of data from the patients, but the potential of these data remains untapped. It retains various relationships and patterns and helps to determine the relationship between clinical parameters like blood pressure, cholesterol, and their association with heart diseases. The objective of the proposed work is to predict the rate of atherosclerosis for a patient based on the clinical parameters like blood pressure and cholesterol and predict the rate using various machine learning classification algorithms. This method will help the medical practitioner to determine the need for an operative procedure based on the outcome. This system predicts the percentage of narrowing down of blood vessels as the output.