ANN based Classification of 3D MR Images of Brain using Frequency filter enhancement

This paper introduces classification of 3D MR images of brain, after enhancement using frequency filtering technique. The proposed method uses voxelwise spatial intensity matrix as features for the ANN classification. The approach consists of three key steps: (1) Convert 3D MRI from spatial to frequency domain (2) Select high frequency components to enhance the image. (3) Again convert the 3D MRI from frequency domain to spatial domain. Experiments are carried out to indicate the high efficiency of the method to classify the 3D MRI from ADNI (Alzheimer’s Disease Neuroimaging Initiative) dataset into AD (Alzheimer’s Disease), MCI(Mild Cognitive Impairment) and Normal.