Article Preview
Top1. Introduction
In the last few decades, cardiovascular disease has become the main cause of death in the world. According to the WHO (World Health Organization), 17.9 million people died worldwide in 2019 from cardiovascular disease, accounting for 32% of all deaths. 85 percent of these deaths are caused by heart attacks and strokes. Since the last three-quarters of cardiovascular deaths took place in low and middle-income countries. 17 million premature deaths are due to no communicable diseases in the year 2019 and 38% of deaths are caused by CVDs. Cardiovascular diseases are high morbidity, high mortality, and high disability in nature. According to the European Society and Cardiology department, in a year nearly 3.6 million people are being diagnosed around the world (Coats, A. J. S., 2019; Spoletini, I., & Seferovic, P., 2017) Most affected by heart disease are from United States (US) (Heidenreich, P. A., 2011). Even with advanced techniques and perfect treatment, 50% of patients cannot fully care for themselves. The popularity of cardiovascular disease becomes a global problem today. Breath shortness, swollen feet, physical body weakness are the common symptoms of heart disease (Durairaj, M., & Ramasamy, N., 2016). Many studies are being conducted to predict heart disease early using various machine learning algorithms and approaches. But the existing techniques are not much reliable in the context of execution time and accuracy. Due to the deficiency of technology and medical experts, heart disease diagnosis and treatment become more difficult. An effective and accurate diagnostic technology can save a large number of patient's life (Al-Shayea, Q. K., 2011). The chances of heart disease patients survive is a maximum of 1-2 years. Generally, the physician checks the patient’s symptoms, physical examination reports, and symptoms for heart disease diagnosis. The outcome of these traditional techniques is not much effective and accurate to detect heart disease in a patient body. However, these approaches are more expensive (Tsanas, A., 2011).