Fuzzy Inference Modeling of Risk Factors in Coronary Diseases : A Review
Abstract
To estimate the variation in the major risk factors for cardiovascular disease (Hemoglobin HGB; mean corpuscular volume MCV; Mean corpuscular hemoglobin concentration MCHC; Fe and Folic acid), we try preventing according coronary heart disease risk factors observed in elderly men and women in the region of Setif – Algeria. Participants.100 men and women aged 26 to 86 years for whom the physiological parameters were recorded. These parameters are risk factors for cardiovascular disease. The expected analysis was estimated using an artificial intelligence model including the principles of fuzzy logic. Risk factors are inputs of the system and the incidence of coronary heart disease is output. The observed data recorded from Analysis Central Laboratory of Setif university hospital - Algeria. Factors that promote coronary heart disease are inaccurate and uncertain. The effect of these factors varies from person to person. Their consideration as fuzzy variables is perfectly adequate. A database is established. Fuzzy inference rules are highlighted according to the recorded values. An algorithmic application is established making it possible to read instantly the number likely the person with a coronary disease just by the random introduction of the variables at the input of the system.
Keywords
Download Options
Introduction
Cardiovascular disease is the cause of death in the industrialized world, and a number of wellcharacterized factors, including advanced age, hypertension, dyslipidemia, diabetes and smoking, contributes to cardiovascular risk1 . Coronary heart disease continues to be a leading cause of adult morbidity and mortality in Europe. Different risk factors are widely studied. However, the weight of each factor varies according to people. The phenomenon is very complex. Modeling such factors by classical mathematical techniques becomes very difficult if not impossible. Several attempts were made. Different models are proposed, but this remains in the realm of probability and approximation.
Nowadays, artificial intelligence has found its application in solving various complex problems. The use of fuzzy logic systems as an intelligent system is a very powerful tool for solving, classifying and making decisions in an uncertain environment, especially in the medical field.
In this study, after giving an overview of the risk factors for coronary heart disease according to the literature, we conclude that these factors are analyzed numerically in all models. In order to get closer to the precision and the expected accuracy, we propose the analysis of these factors by the techniques of artificial intelligence in particular the principles of fuzzy inference. For this purpose, we give a general overview on the fundamental notions of fuzzy logic in order to facilitate the understanding of its application.
Conclusion
Once the rule base is established, it becomes possible to instantly read the degree of coronary heart disease (Figure 4). The result is the collaboration of the set of rules that support all input variables. Since the input variables are considered as fuzzy variables by expressing them by linguistic variables, this gives an analysis as precise as possible. Also the output variable is expressed in linguistic terms concerning the incidence of coronary disease. This also gives the possibility of reading a result in a wide numerical and symbolic range.
At the end, the resulting application makes it possible to randomly display values at the input to read the result at the output. If all of the factors are taken in a precise manner and the rules are established correctly and encompassing all possibilities, it becomes possible to predict the onset of coronary heart disease without making appropriate diagnoses. This tool can be considered as an aid to doctors in their diagnosis, prevention and treatment of coronary patients.