|
Abstract
|
Mathematical modeling is one of the feasible methods that can be used to solve real problems. The paper focuses on different mathematical and statistical approaches to prediction disease. Prediction is essential on some issues such as in health. Among the various disease in the field of health, Stroke is as the third leading cause of death and the first major cause of human long-term disability, considered by researchers. The main goal of this paper is to design and construct a prediction model for the patient, based on clinical reports, either the stroke will happen or not. In this way, considering the kind of collected data, we apply the logistic regression method for select inputs. After selected sample of large data set, we examine the capability of Data Envelopment Analysis to predict occurrence or non-occurrence of Stroke. The results showed that if any patient was on the frontier, he/she did have Stroke by one risk factor. On the other hand, if observation belong on the below of the frontier, they are “Non-Stroke” or “Stroke with more than the one risk factor”.
|