Development of Computer-aided Classification of Alzheimer's Disease Based on Cerebral Blood Flow map by MR imaging

The aim of this study was to develop a computer-aided classification system for AD patients based on functional imaging features derived from the cerebral blood flow(CBF)maps measured by arterial spin labeling(ASL)technique which is one of MR imaging techniques. In the first step, the average CBFs in 16 cortical regions were determined as functional imaging features based on the CBF maps, which was non-linearly registered to a Talairach brain atlas by using a free-form deformation tecunique. In the second step, a support vector machine was trained by the average CBFs in 6 cortical functional regions which are known as hypoperfused regions(decreasing regions of blood flow)in AD, and then was employed to distinguish patients with AD from control subjects. For evaluation of the proposed method, we applied it to 15 AD patients and 15 control subjects. As a result, the area under the receiver operating characteristic curve was 0.903 based on a leave-one-out-by-case test. Our preliminary results suggest that the proposed method based on functional imaging features obtained by the ASL technique would be feasible for detection of patients with AD.