Heart Disease Classification Using Regional Wall Thickness by Ensemble Classifier

Abstract

In recent years, heart disease is becoming a major problem in human beings. The diverse of syndromes that will infect the heart is known as the heart disease. Cardiac magnetic resonance images are formed using the radio waves and an influential magnetic field, which will produce pictures with a detailed structure of within and around the heart which can be used to identify the cardiac disease through various learning techniques that are used to evaluate the heart's anatomy and function in patients. In this chapter, ensemble classification model is used to classify the type of heart disease. Automated cardiac diagnosis challenge dataset is taken for prediction of heart disease that consists of 150 subjects which is evenly divided among all five classes. The dataset is initially pre-processed to eliminate the noise in image followed by the Region of Interest extraction and segmentation based on densely fully convolutional network, and the feature extraction to extract the values to calculate the ejection fraction value. Based on the Then, the heart disease is classified by using the ejection fraction value representing the regional wall thickness.