Font Size: a A A

The Research On Individual Brain Morphological Network Based On Mild Cognitive Impairment

Posted on:2021-06-20Degree:MasterType:Thesis
Country:ChinaCandidate:M LvFull Text:PDF
GTID:2504306470473634Subject:Medical imaging and nuclear medicine
Abstract/Summary:PDF Full Text Request
Objective:The research focused on the construction of individual brain structural network using the method of voxel-based morphometry and Kullback–Leibler divergen,and explored the differences of the brain topological properties among the healthy control aging,MCI-to-AD converters and MCI-to-AD non-converters.Subjects and methods:To estimate brain topological properties,we first screened the quantitative MRI scans of 529 subjects from public ADNI(Alzheimer’s disease Neuroimaging Initiative)database,including the subjects with MCI who have converted to AD after 9-year(cases: n=173),and the subjects with MCI who have remain stable after 9-year(cases: n=131),and the subjects who have remain healthy aging after 9-year(cases: n=225).Secondly,the structural data-preprocessing and the calculating of gray matter volume(GMV)were performed by the VBM toolbox in SPM software.The probability density distribution of GMV was performed for each brain region based on the priori brain atlas(AAL-246 mask).According to the probability density distribution,the morphological similarity between brain regions was measured using KL divergence for obtaining the individual brain network construction.Finally,the brain topological properties were calculated using the method of graph theory under the different spasity.Results:Compared with the healthy controls,the increased nodal betweeness of parahippocampus gyrus was found in the subjects with MCI who have remain stable and the increases were found in their nodal degree and efficiency of bilateral occipital lobe and right hippocampus.However,the reductions were found in the path length of left occipital lobe and right anterior hippocampus in the ncMCI subjects.In addition,the ncMCI subjects represented the increase of path length and the decrease of nodal degree and efficiency in right middle frontal gyrus.Compared with the healthy controls,the increased nodal betweeness of bilateral posterior hippocampus,right anterior hippocampus and left thalamus was found in the subjects with MCI who have converted to AD.However,the reductions were found in the nodal betweeness of parahippocampus gyrus and left amygdala in the cMCI subjects.In addition,the cMCI subjects represented the increase of nodal degree and efficiency and the decrease of path length in the bilateral hippocampus,bilateral occipital lobe,left amygdala and left thalamus.The cMCI subjects also represented the decrease of nodal efficiency and the increase of path length in the bilateral superior temporal sulcus,right middle frontal gyrus and right middle temporal gyrus.In the cluster coefficient,the increase of the bilateral parahippocampus,bilateral amygdala and left thalamus and the decrease of left posterior hippocampus were found in the cMCI subjects.In the local efficiency,the increase of the bilateral parahippocampus,bilateral amygdala and left thalamus and the decrease of left posterior hippocampus were found in the cMCI subjects.Compared with the ncMCI subjects,the increased nodal degree and efficiency of bilateral anterior hippocampus were found in the subjects with cMCI subject and the decreases were found in the nodal path length of these regions.Also,the increase of the cluster coefficient and local efficiency was found in bilateral para hippocampus and left amygdala of the cMCI subjects.Conclusions: In this research,we found individual structural network based on regional morphological similarity among cMCI,ncMCI and NC group,and investigated significant difference in the cMCI and ncMCI groups compared with NC group.Moreover,the abnormality of topological properties mainly located in bilateral parahippocampus,anterior hippocampus,middle temporal gyrus and left amygdala in the cMCI subjects,indicating the topological properties of these regions may predict further degeneration of the cMCI subjects.
Keywords/Search Tags:Individual Structrual Network, Kullback–Leibler divergen, VBM probability density function, MCI
PDF Full Text Request
Related items