Font Size: a A A

Research On Brain Networks For Human Language Semantic Mechanism Based On FMRI

Posted on:2016-03-20Degree:MasterType:Thesis
Country:ChinaCandidate:Q LiuFull Text:PDF
GTID:2284330479998999Subject:Biomedical engineering
Abstract/Summary:PDF Full Text Request
The study of brain networks is the forefront exploration of neurological diseases diagnosis and rehabilitation project. Research on brain networks based on the fMRI is expected to interpret the working mechanism of a brain tissue on the basis of blood oxygenation level-dependent(BOLD) functional imaging. This study aims at exploring the active brain regions of normal adults according to the language task. It’s hopeful to provide foundations of clinical preoperative localization and nerve recovery for aphasia and temporal lobe epilepsy disease.The data adopted in this thesis is the third group data from the human brain connection group project(HCP) data of America which is published in the autumn of 2013. The experiments are about language task for 10 individuals, and their resting state data for comparison. The preprocessing is done firstly, and then the regional homogeneity(ReHo) method and the ‘small-world’ network analysis method are applied to analyze the two sets of data, to locate the language functional areas of the brain involved in the process of semantic processing.The small-world network analysis method is used to construct the brain functional network. The difference of the functional networks between language task state and resting state is compared according to the small-world topological perspective. Compared with the resting network, the node degree, clustering coefficient and average path length of the language task network has changed, and it satisfies the small-world property. With the same average of node degree, the results show that the language functional network has smaller clustering coefficient and shorter average path length than the resting functional network.The regional homogeneity(ReHo) method is used to analyze the two sets of the data. In the two-sample T-test, language tasks can significantly activate Broca’s area and Wernicke’s area. Compared to the resting state, activation results show that the regions with enhanced ReHo value are distributed in superior temporal gyrus and middle temporal gyrus, orbit inferior frontal gyrus, middle frontal gyrus in both hemispheres and triangle of inferior frontal gyrus in left hemisphere; and the regions with reduced ReHo value are in angular gyrus, inferior parietal lobule, triangle of inferior frontal gyrus in both hemispheres, inferior frontal gyrus in right hemisphere. In semantic judgment tasks, the frontal, top, pillow, temporal lobe, and other brain regions of the hemispheres are presenting activation of different degrees, which indicate that the generation of any function requires the synergy of brain.
Keywords/Search Tags:brain network, functional magnetic resonance imaging(fMRI), regional homogeneity(ReHo) method, small-world network, language function
PDF Full Text Request
Related items