Most computer-aided diagnosis systems for Alzheimer's disease (AD) using magnetic resonance (MR) imaging were based on morphological image features, not functional image features, which would be also useful for diagnosis of AD. The aim of this study was to develop a computer-aided classification system for AD patients based on functional image 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 ten cortical regions were determined as functional image features based on the CBF map image, which was nonlinearly registered to a Talairach brain atlas. In the next step, a support vector machine was trained by the average CBFs in ten cortical functional regions, and was employed for distinguishing patients with AD from control subjects. For evaluation of the method, we applied the proposed method to 20 cases including ten AD patients and ten control subjects, who were scanned at a 3.0-Tesla MR unit. As a result, the area under the receiver operating characteristic curve was 0.893 based on a leave-one-out-by-case test. The proposed method would be feasible for classification of patients with AD.