Research On Brain Structure Network Of Alzheimer’s Disease

Alzheimer’s disease (Alzheimer’s disease, AD) which is a neurodegenerative disease oftenoccurs in the elderly, while mild cognitive impairment (MCI) has a high risk for developinginto AD. Unfortunately, there has no effective ways for MCI/AD treatment. Therefore, theearly detection, diagnosis and prevention of MCI and AD, postpone the proceeding is crucial.However, at present, the pathogenic mechanism and process of AD are still unknown, andthere hasn’t any objective and effective biological marks for early prediction and diagnosis ofeither MCI or AD. In this study, the neuroimaging techniques are applied to compare andanalyze differences of brain structure among AD, MCI and normal aged (normal control, NC),explore sensitivity characteristics of MCI/AD, so as to provide potentials for early detectionand diagnosis of MCI, AD.First of all, we apply the voxel-based morphometry(VBM) method to compare thedifferences of gray matter in brain regions among NC, MCI and AD groups. Then, weconstruct brain structure networks based on graph theory, and investigate the connectionsbetween brain areas, including the modularity, core node, and quantitative parameters of brainnetwork, such as small world property, clustering coefficient, shortest path length. Finally, thenetworks of gray matter of three groups are compared and analyzed using qualitative andquantitative measurements.Experimental results show that MCI suffers gray matter atrophy in brain areas such ashippocampus, parahippocampal, olfactory, amygdala, temporal lobe, moreover, for AD group,gray matter atrophy occurs in much more brain areas related the default network, frontal lobe,occipital lobe etc. From NC to MCI and then to AD, the connections between memoryassociative system, visual, auditory and motion system are gradually weakening. and the brainregions involved are gradually increasing. However, the connections between occipital lobeand other brain regions are gradually increasing. The NC group has the best performance inglobal network efficiency and network topology, the worse is MCI group and the worst is ADgroup.In addition, we also carry out a preliminary investigation the changes of MCI patients, inother words, we make the longitudinal comparison about the gray matter and brain structurenetwork on MCI, MCI2AD(MCI converting into AD was confirmed at the first time) andMCI2AD_latest(the latest images of MCI2AD group). From MCI to MCI2AD andMCI2AD_latest, the trend of gray matter atrophy of brain areas is deepening and expanding,the global efficiency, topology are worsen, but local efficiency improve. The betweenness centrality value of cingulum_mid of the three groups decrease gradually. the integrality of thedefault network is also gradually destroyed. Compared to MCI, MCI2AD_latest group hastwo hubs in occipital lobe, one is occipital_mid_R, the other is cuneus_R, It is demonstatedthat the betweenness centrality value of brain areas in occipital lobe increase in AD groups,which is consistent with the above comparison among NC, MCI and AD group.