Attention Deficit hyperactivity disorder (ADHD), which is one of the most common abnormalities, usually appears at the early years of life. ADHD is usually observed with some structural changes at different regions of the brain cortex. Exploring those structural variations can have an important role in diagnosis and treatment of disease symptoms. In this paper, we propose a new framework for statistically exploring the main volume changes of the brain cortex in a group of 50 ADHD (combined-type) patients compared to controls. For this purpose, all the images are primarily registered with the ICBM152 template. Then, the gray-matter region of each image is segmented to obtain its corresponding probability map. To compute the local concentration/volume, the gray-matter probability map is smoothed by using a Gaussian filter with enough-large standard deviation. Next, through the hypothesis testing algorithm, the voxels with the significant volume change in all images of the ADHD group compared to controls are detected. Finally, the clusters with enough significant voxels are extracted as the regions with the significant volume changes due to ADHD. Finally, the exact position of every extracted cluster is obtained by using the parcellation atlas AAL2. Experimental results demonstrated significant volume increase in leftt medial frontal lobe, right cuneus, right rectus, right medial temporal gyrus, and left lingual gyrus. Also, there was significant volume decrease in right calcarine.

Keywords