ADVANCED DETECTION OF BLOCKED CORONARY ARTERIES USING MACHINE LEARNING ALGORITHMS

Sujesh Shankar Dinesh Mandal

Abstract: The healthcare industry is observing a tremendous advancement along with upcoming innovations in Information Technology and Computer Science. A common task in Machine Learning is to classify data. An essential task for extracting knowledge from large databases is done by Data Mining. Data Mining in the Healthcare industry is an upcoming field of interest not only for Data Scientists but for Medical Experts for providing deeper understanding and prognosis of medical data. A majority of data mining methods depend on a set of features that define the learning algorithm directly or indirectly and influence the complexity of the resulting models. In the last 10 years, heart disease has been the leading cause of deaths in the world. A lot of researchers have been using data mining techniques to diagnose heart diseases using machine learning algorithms. Coronary Artery Disease [CAD] is a chronic disease that occurs when there is usual cause is the buildup of plaque. This causes coronary arteries to narrow, limiting blood flow to the heart. Coronary artery disease can range from no symptoms, to chest pain, to a heart attack. Treatments include lifestyle changes, medication, angioplasty and surgery. To reduce the large scale of deaths from Coronary Artery Disease, an efficient and quick detection technique is to be researched. Various steps are taken to handle the outburst of information related to medical sciences and acquisition of valuable knowledge. Data Analytics and Machine Learning Algorithms play a vital role in this area. This paper presents the Naïve Bayes algorithm, Decision Tree Algorithm using Entropy function, Support Vector Machines and Logistic Regression algorithm for early detection of blocked coronary arteries. The purpose is to enhance the accuracy and enhance the flexibility of the algorithm.

Keywords: Data Mining, Coronary Artery Disease, Machine Learning, Naïve Bayes Method, Support Vector Machines, Big data analytics and Predictive analytics

DOI: https://doi.org/10.15623/ijret.2018.0707026