This study proposes a computer-based decision support system to investigate the distinctive factors of diabetes mellitus (DM) with ischemic (non-embolic type) stroke and without stroke. Database consists of a total of 16 features that are collected from 44 diabetic patients. Features include age, gender, duration of diabetes, cholesterol, higher density lipoprotein (HDL), triglicerit levels, neuropathy, nephropathy, retinopathy, peripheral vascular disease (PVD), myocard infarction (MI) rate, glucose levels, taking medicine, blood pressure. Metric and non-metric features are distinguished. First, the statistics, mean and covariance, of data are estimated and the correlated components are observed. Second, principal component analysis (PCA) is used for major components. Finally, decision making approaches, k-nearest neighbor (k-NN) and MLP, are employed for classification of all the components and major components case. Macrovascular changes emerged as principal distinctive factors of ischemic-stroke in DM. Microvascular changes were generally ineffective discriminators. Recommendations were made according to the rules of evidence-based medicine. Briefly, this case study supports theories of stroke in DM and also concludes that the use of intelligent data analysis improves personalized prevention.