Using Fuzzy Linear Regression Models to Identify the Factors Affecting Myocardial Infarction

This study focuses on the application of many fuzzy linear regression models to analyses medical data concerning myocardial infarction, particularly with the level of the troponin, which is vital for identifying heart attacks. More specifically, the physiological indicators describing the state of the patient including blood pressure, blood sugar, creatinine, and the relationship between these indicators and the heart attack enzyme level.To accomplish the aim of study, the necessary data were obtained from the hospitals that treated patients with a diagnosis of myocardial infarction. Then the performance of many fuzzy regression methods was compared employing these data.The study shows that the fuzzy least squares procedure has the lowest mean squared error values of all the models, and the best accuracy in simulating the effect of physiological factors on the level of the heart attack enzyme. Also, this study emphasizes the importance of applying fuzzy regression in medical statistics due to the existence of uncertainty in the field and the applicability of this method to enhance predictive power and decision-making in healthcare.