Heart-Health Status Using Machine Learning
- June 29, 2021
- Posted by: rsispostadmin
- Categories: Computer Science and Engineering, IJRIAS
International Journal of Research and Innovation in Applied Science (IJRIAS) | Volume VI, Issue V, May 2021|ISSN 2454-6194
Heart-Health Status Using Machine Learning
Ebenezer Olukunle Oyebode
Computer Science Department, Ajayi Crowther University, Oyo, Nigeria
Abstract: Heart disease is one of the killer diseases in the world. Early detection of the disease is one of the ways to salvage affected people. The use of machine learning techniques can be used to offer solution to the detection of heart diseases. In this study the accuracy of prediction of some tools of machine learning has been carried out. The performance evaluation of the three models have been carried out using precision, recall, F1-score and accuracy. The results obtained showed that Logistic regession model out performed others in terms of precision, recall, F1-score and accuracy.
Keywords: machine learning, recall, F1-score, precision
1.INTRODUCTION
Heart disease is one of the killer diseases that affect humanity. World Health Organisation (WHO) confirmed that about twelve (12) million deaths occur annually through heart failure across the world. About 26 million people are having heart diseases world-wide and the number of affected people can be on the increase except right precautions/steps are taken to salvage the situation [6]. Coronary Heart Disease (CHD) and myocardial infarction (as heart attack) are variation of the heart diseases. Risk factors that can contribute to heart problem include smoking, drinking, bad weather, poor diet and dirty environment among others [4]. Features that manifest the presence of heart diseases include high level of cholesterol, high blood pressure, high pulse rates among others.
Machine learning has been applied in health sector as a tool to complement for diagnosis of diseases and yield improved results [2]. Machine learning techniques in recent times have witnessed significant progress in terms of clinical researches. Even though developing machine learning algorithms require significant amount of time and understanding of how the underlying algorithms work, machine learning algorithms can be used for detection of heart diseases.
