Cortical Thickness Brain Network Analysis Based On Alzheimer’s Disease

Alzheimer’s disease (AD) is the most common form of dementia among elderly people and seriously affects the health of older people.Mild cognitive impairment (MCI)is a concept that used to describe a high-risk "pre-demen-tia" state.A desirable aim of AD research is to develop means of making early and accurate diagnoses of individuals who either have or are at increased risk of developing AD.This will allow for timely interventions that may be effective in managing or treating AD.However, currently there are no accurate and effective biomarkers for early diagnosis of AD. The main aim of this study is to explore the theroy and method of brain structural network about MCI and AD.Here cortical thickness was used as morphological indicator to construct structural network, then we extracted network connection and propertis as features to research on the early prediction and imaging biomarkers for the development of AD.Based on neuroimaging and genetics, this paper investigated the pathomechanism of AD to offer some contributions on early precise detection and diagnosis of AD.The main contents and contributions of the study are as follows:1)We classify and process the magnetic resonance imaging (MRI) data firstly. The progression of MCI varies enormously among the people; some revert to normal status, some remain stable for many years which is defined as stable MCI (sMCI), and other convert to AD rapidly which is defined as converted MCI (cMCI).We investigated corss-sectional and longitudinal changes of the cortical thickness of normal control (NC), sMCI, cMCI and AD subjects.We found that the shrinking area spread from the default mode network (DMN)to the whole brain.Then, we quantify the atrophy rate of each brain region which is significant for the the prediction and diagnosis of AD.2)Constructing the brain structure networks based on the cortical thickness using graphics principles to find the corss-sctional and longitudinal differences of network properties among NC. sMCI.cMCI and AD patients, so as to provide reliable classification characteristics of the four groups.Fully utilizing the longitudinal information of all groups, a sigmoid dynamic pattern of small world characters can be detected as the hypothetical continuous longitudinal stages of AD development, that begin from normal elderly to AD patients via MCI stage.On one hand, the results provide a potential explaination for the previous inconsistent studies.On the other hand, the methods and the severity of disease may have effect on the network comparision results, which provide new evidence for understanding the pathophysiology and abnormal network of MCI and AD.Last but not least, referring to the underlying mechanism of network property, whether it can be used as a biomarker of AD remains to be examined.