Rating and Comparison of Well-known Cardiovascular Risk Factors in Kermanian Male Patients Using Fuzzy Linear Regression

Cardiovascular diseases are still among the most important causes of death in different countries. There are several risk factors for the onset of this disease. The rating of these risk factors is very important for informing the community and planning for the future.
Linear regression is one of the classic statistical methods that has many applications in medical sciences. When dealing with fuzzy data, it is not possible to use linear regression. The use of angiography to estimate the extent of congestion is associated with an estimate of more or less stenosis and the increase of atherosclerosis. Hence, this variable has been considered as a fuzzy variable. Fuzzy linear regression (FLR) was used to rank and compare the known risk factors for coronary artery occlusion.
After analyzing the data by least squares FLR, the most important risk factors included Family history, history of diabetes, age, history of hypercholesterolemia, history of cigarette smoking, Body Mass Index and history of hypertension respectively.
When it is not possible to calculate the response variable or one of the independent variables examined in the model accurately, FLR, or to be more precise, regression in fuzzy environment can be a good alternative for conventional regression analysis.