A study on analysis of cardiovascular diseases

Abstract

Commonly seen in adults, cardiovascular diseases are important health problems. In order to investigate the causes of the diseases which affect the heart and the blood vessels, two datasets were used. First, one of these datasets is publicly available dataset provided by the University of California, Irvine Machine Learning  Repository. The effects of  biochemistry and hemogram laboratory test results for the Cardiovascular Diseases were analyzed by using the second dataset which was taken from the cardiology and other services of Yildirim Beyazit University Ataturk Training and Research Hospital. ICD-10 (International Statistical Classification of Diseases and Related Health Problems) booklet was taken as a reference for the patient and control groups. The successes of the classifier algorithms indicated that working with the datasets which have only limited number of attributes is not right step.

Keywords: Cardiovascular diseases; logistic regression; machine  learning; medical data; random forest.