Different Deep Learning Techniques in Heart Disease Classification: Survey

Abstract

Cardiovascular disease prediction is a serious challenge for clinical data analysis. This study examines deep learning-based categorization strategies for heart disease. Deep learning algorithms are employed with echocardiograms to categorize heart disease. This paper uses echocardiography to predict and identify heart abnormalities, with the help of decision-making and forecasting, based on the copious data the healthcare sector has provided. Medical experts can forecast clinical outcomes, which helps them choose the best course of action. In-depth longitudinal electronic health records are a rich source of historical data with complex patterns that have the potential to be leveraged by machine learning to improve physicians' prediction abilities (EHR) significantly. Most contemporary medical specialties rely on imaging as one of the most data-rich components of electronic health records when making treatment decisions (EHRs). Only a few tasks in medical image processing and reconstruction have seen success using machine and deep learning, including registration, segmentation, and feature extraction. Cardiac imaging sequence analysis must incorporate the extraction of spatial and temporal characteristics to forecast crucial information throughout time correctly.