Development of an Intelligent Model to Estimate the Probability of having Metabolic Syndrome

详细信息    查看全文

• 作者：Nuriye Sancar^a ; ^{nuriye.sancar@neu.edu.tr" class="auth_mail" title="E-mail the corresponding author} ; Mehtap Tinazli^b
• 关键词：Metabolic syndrome ; risk factors ; logistic regression ; odds ratio
• 刊名：Procedia Computer Science
• 出版年：2016
• 出版时间：2016
• 年：2016
• 卷：102
• 期：Complete
• 页码：106-112
• 全文大小：137 K

文摘

Logistic regression has now become an essential part of medical data analysis that uses a binary-response model. The model is frequently used by epidemiologists as a model for the probability (interpreted as the risk) that an individual will acquire a disease during a specified time period, during which he or she is exposed to a condition (called a risk factor) known to be or suspected of being associated with the disease. The objective is to establish a model using a minimum number of variables, and is also able to identify the relationship between the dependent variable and independent variable. Additionally, the study will determine the risk factors that can lead to the development of metabolic syndrome (MetSyn) and will establish an intelligent and biologically acceptable model for estimating the probability of having the condition, based on the NCEP ATP III criteria. In this study, binary logistic regression analysis has been employed in order to specify the risk factors that affect metabolic syndrome. Metabolic syndrome (MetSyn) is a common metabolic disorder that is increasingly caused by the pervasiveness of obesity in society and diagnosed according to the National Cholesterol Education Program (NCEP) Adult Treatment Panel III (ATP III) Identification¹. The data has been obtained from the laboratory test results of 321 adult individuals who had consecutively been treated by the Near East University Internal Medicine Department. For this intelligent model, binary logistic regression analysis has been used. The sensitivity, specificity and accuracy rates have been detected as 94.7%, 96.0% and 95.5%, respectively. As a result, homeostatic model assessment (HOMA-IR), uric acid, body mass index (BMI), low-density lipoprotein (LDL) cholesterol, age, smoking, education level (EL) are defined as metabolic syndrome risk factors, the model has been estimated by using those variables in the acquired intelligent model. As a consequence of the research, it has been determined that the key elements that can have an impact are the changeable risk factors, meaning that the illness could be destroyed before it actually occurs, and lifestyle change, that can also prevent the illness.