| The traditional f MRI data research of MCI mainly focuses on static functional connection.The implicit assumption of static functional connectivity is that the connections between brain regions are constant during scanning.However,studies have shown that there are dynamic neural interactions in the resting brain,and the resting brain is not constant.In view of the shortcomings of static function connection methods,dynamic function connection methods are widely used in recent years.Effectiveness of dynamic functional connectivity approach demonstrated in studies on the aging brain,parkinson’s disease and depression.Due to the current lack of research on the dynamic network of MCI,this paper compared the differences between healthy subjects(HC)and MCI dynamic and static brain networks based on the data-driven method of independent component analysis(ICA)and the model driven method of brain atlas(AAL).The results are as follows:(1)In the static network,we found that there are many differences in the brain network regions of MCI and HC,among which the main differences are the default network(DM)and the cognitive control network(CC).The reciprocal connections of HC significantly stronger than that of MCI are mainly distributed in CC,DM and SM,and the reciprocal connections of MCI significantly stronger than HC are distributed in almost seven network regions,indicating that there may be a functional compensation mechanism in the brain of MCI.(2)Two highly structured states of the brain were found in the dynamic functional connectivity network,The first state occured frequently and the network was weakly interconnected(weak interaction state),and the second state occured frequently and the network was strongly interconnected(strong interaction state).The state characteristics of MCI and HC were compared and analyzed experimentally,and it was found that the number of state transitions of MCI was significantly lower than that of HC,and the dwell time of MCI weak interaction state was significantly higher than that of HC weak interaction state.(3)In the dynamic and static function connection networks,the two groups of subjects had differences in topological attributes in multiple regions,most of which were distributed in the right brain,and the frontal lobe was the main focus of the differences.Compared with static function connection networks,more reliable topological attribute differences can be found in dynamic functional connection networks.The small world attribute of MCI was lower than HC,and the network compatibility was higher than HC.The right brain network of MCI was seriously damaged,and the frontal lobe mainly undertaked functional compensation and remodeling.The significantly different network topology attributes could provide data reference for MCI diagnosis. |
PDF Full Text Request