| There are obvious differences between male and female in personality and behavior.For example,female is more emotional than male,is better at language expression,and has better short-term memory skills,while male is more mature and decisive than female,is better at rational thinking,and has higher athletic talents.Previously,most of the research on these differences were explored by constructing a group-level morphological brain network,but such methods may mask subtle structural differences of interest to the clinical population.Therefore,this thesis will study the similarities and differences of brain morphology and network between the two sexes at the individual level,so as to help us have a better understanding of the differences between the two sexes in thinking and behavior.The research object of this thesis is the structural data of 343 normal people(176 females and 167 males).The main research contents are as follows: Firstly,the brain structure data images of the participants were preprocessed to obtain smooth gray matter,white matter and Cerebrospinal fluid images.Then the differences in total gray matter volume,total white matter volume,the proportion of gray matter volume and the proportion of white matter volume between the male group and the female group were compared using the two-sample t-test.Then,the90 regions of interest(ROI)of the Anatomical Automatic Labeling template(AAL)were defined as nodes,and the probability distribution function of the gray matter volume values of all voxels of each ROI was estimated by using the kernel density estimation automatically selected by the bandwidth,and the similarity measure of the two ROIs was calculated and defined as the connection weight to construct a morphological brain network for each participant.Then the network based statistic method was used to find the sub-networks with significant differences between male and female groups.Then the mean value of the different sub-network connections was taken as the feature to construct the support vector machines for classification and discrimination of male and female subjects.Finally,the network topological properties were used to analyze the efficiency of information flow in the brain of male and female,and to explain the differences between male and female in language,movement,emotion,logic and memory performance.Finally,this thesis got the three main results: First,when the brain volume changes together,it was found that the proportion of gray matter in the female brain was higher,which may be due to the dense distribution of female neurons,leading the female to bleed faster,to have a shorter cure time to certain medications than male.Second,two sub-networks with significant differences between male and female was found through the network based statistic method.One of the sub-networks mainlycontained connections between the caudate nucleus and other nodes,all of which were higher in female than in male.Another sub-network was the opposite.female were lower than male,mainly including the parietal part,the posterior cingulate and the insula.Then,this thesis adopted the connection mean of these difference networks as the feature to construct the support vector machines,which found that the classification accuracy of male and female who had been tested was as high as 96%.Third,the network topological properties were used to analyze the global and local information flow efficiency of male and female brains,which found that female was better than male in language,emotion and memory,while male was better than female in thinking logic,movement and direction.This thesis found that there are differences in individual morphological brain network's connections and graph properties between the brains of male and female,and those differences can be considered to be related to the differences in brain areas which is responsible for relevant functions between the two sexes,helping us have a better understanding of the differences between the two sexes in thinking and behavior. |
PDF Full Text Request