Using Discriminant Analysis for Classification of Patient Status after Three Months from Brain Stroke

Abstract

Brain Stroke is a medical condition in which poor blood flow to the brain results in cell death. It also effects in parts of the brain not functioning properly. This situation is one of the main cause of death and a leading cause of disability in Kurdistan region and in worldwide. The aim of this paper is to predict the membership in two mutually exclusive groups (Group Died and Group Alive) after three months from brain stroke by using discriminant analysis which has used to classify the status of patient. The data of the research has been collected in Rzgari Hospital and Rojhalat Emergency Hospital of Erbil. The questionnaire form consists of 29 variables, one of them dependent and other variables are independent appendix (1). Sample size of the research contains 134 patients which faced brain stroke.  In the practical part of the research we have built two models based on variable gender and model of both gender together by using stepwise method of discriminant analysis. After creating the three models we compared the results of the classification. The classification results show that the percentage of patients classified correctly for male and female separately higher than the percentage of patient classified correctly for the model of both genders together. For analysis, (SPSS V25) was used.