Automatic classification of Alzheimer´s patients and age-matched healthy subjects using independent component analysis

In this paper, we proposed an automated novel method combining independent component analysis and voxel of interest for classification of MRI scans of Alzheimer patients and healthy subjects. The method mainly includes five steps: preprocessing of MR images, segmentation of gray matter of the brain, decomposition using independent component analysis, extraction of voxel of interest, and classification by a support vector machine classifier. The automated method of selection of voxel of interest has no disadvantages of manual voxel of interest, such as labor-intensive and time-consuming. The experimental results indicate that the proposed method is able to provide better classification results than that in some previously related works.