Brain Connectivity Network In Autism Using Resting State Functional Magnetic Resonance Imaging

Functional magnetic resonance imaging (fMRI) technology is a non-invasive brain imaging technique that can quickly reflect the functional activity in the brain.The brain activity signals in resting state have important physiological significance. Autism is a neurological disorders with high incidence. It is short of a better treatment and its diagnosis is subjectivity. Therefore, In this study, we investigated the topological property of the brain functional connect network between autism patients and normal control. It can reveal the differences between autism and normal in brain function network, and it can look for relevant regional dysfunction. This will have important implications in the clinical and scientific aspects of the brain.fMRI were gathered from30autism patients and30normal adults. They has no significant different in age or gender. We preprocessed the resting-state fMRI data, and then registrated the image to standard space and extract time series signals of interest brain areas.First, we constructed correlation matrices between all the108brain regions, and then threshold the correlation matrices to build undirected binary graphs for each subject and compared topological property of brain functional networks between the two groups by applying graph theoretical approaches. Through the global brain network analysis, we got25regions which shown significant differences between the two groups. Further we compared topological property of local brain functional networks between the two groups. Finally, we compared the default network topology properties between autism and normal control.We found that the global network topology properties showed significant changes in autism. The change of local brain network topology properties is similar to that of the global brain network. Compared with normal control group, the global efficiency of autism are higher than those of the control group, other graph metrics were lower. In addition, the local network analysis also found some brain regions which had significant difference in patients with autism. They are Superior Frontal gyrus, Orbitofrontal cortex, Middle Occipital gyrus and so on. The default network in autism also has a weaker functional connectivity. Cluster coefficient, characteristic path length, local efficiency and betweenness centrality are also lower than normal control.In summary, we found that the brain network integration decreases and brain network segregation increases in the autism group compared with the healthy control group. The abnormal in brain function network will help us to understand brain dysfunction in patients with autism.