Fundamentals and Methods of Machine and Deep Learning

8Machine Learning and Deep Learning for Medical Analysis—A Case Study on Heart Disease Data

Swetha A.M., Santhi B.* and Brindha G.R.

SASTRA Deemed University, Thanjavur, Tamil Nadu, India

Abstract

Cardiovascular diseases (CVDs) encompass a variety of heart problems ranging from vascular disorders like coronary heart diseases and peripheral arterial disorders to morbid cardiac diseases like heart failure (or myocardial infarction) and cardiomyopathy, to name a few. These diseases often occur as repercussions of low cardiac output and decreased ejection factor, usually exacerbated by vascular blockages. With the increasing severity of CVDs, a need for predicting heart failures is on the rise, but the traditional methods employed for CVD-related event prediction, unfortunately, have failed to achieve the acme of accuracy. Given a set of medical records as datasets, Machine Learning (ML) can be employed to achieve high accuracy in the prediction of patient survival and also in determining the driving factors that increase mortality among CVD patients. The medical records that provide the necessary data for prediction form a basic framework that divulges inconspicuous consistencies of patient’s data which along with an appropriate ML algorithm confute the traditional methods thereby providing a strong base for determining the feature that contributes the most for the risk factor. The proposed model uses various feature selection techniques to extract those features in particular ...