Study And Modeling Of Resting-state Brain Functional Network Based On Complex Network

Human brain is known as the most complicated and excellent system of dynamic information processing in nature at present. It can accomplish a series of complex functions by tens of millions of neurons continuously organizing and re-organizing its functional connectivity. Apply the theory and methods of complex network to neuroscience, which is very important for helping human to perceive and comprehend the behavior of brain, and at the same time using systematic views to analyze the information processing and transmission of brain mechanism. The present study based on the functional magnetic resonance imaging and applied the theory of complex netowrk to construct and analyze the brain functional nework of the normal and the schizorphrenia in resting state. Then we researched the structure and function of brain and discussed the characteristics of brain network in normal and abnormal physiological. The main contents and contributions of the dissertation are as follows:The research of brain functional connectivity was made as a core, the backgroud knowledge of the development and principle of functional magnetic resonance imaging and the theory of complex network were introduced briefly. Then, all kinds of prcessing methods of fMRI data were minutely recommended in the point of view of function segregation and function integration. The number of stages and significance of statistical parametric mapping as a classic method of data preprocessing and the methods of functional connectivity were explained in detail.Combined the theory of complex network, we constructed and analyzed the model of the normal human brain network, and got some meaningful results. When the correlation analysis of seed was used to construct adjacency matrix, the selection of threshold value need following the two basic principles, that is, the integrity of networks and the small-world properties, to avoid the over voluntary of threshold and make the brain network most represent the characteristics of the actual brain system. On this basis, we analyzed the characteristics of brain functional connectivity and computed the centrality indices. The results showed that the resting-state brain functional network was a scale-free small-world network and the posterior cingulum, precuneus, cuneus and superior parietal lobule played important roles.For the high complexity of the resting-state fMRI data of schizorphrenia, the brain functional network was constructed and all kinds of its characteristics were analyzed based on the level of voxel. We got the result that the brain network of schizorphrenia also has the scale-free small-world characteristic, it is correspondent with the research in the perspective of brain regions. Further, compared the brain functional network of the normal and the patients with schizorphornia, the outcome showed that the characteristic of small world decreased to some extent. The study also displayed that the model of complex network is a very useful tool for detecting and estimating the brain functional network, and contribute to understand the working mechanism of brain in the state of abnormal physiological conditions. At last, we concluded the work of paper and summarized what we could do next in the research of brain functional connectivity form the overall trend.