Early Diagnosis Of AD Based On Lateralization Of Structural Brain Network

Alzheimer’s disease(AD),a common neurodegenerative disease,is one of the most common causes of cognitive impairment in the elderly.The onset of AD is insidious and the incubation period is long,which makes the clinical diagnosis very difficult.Once the disease starts,the brain injury cannot be reversed.At present,the pathological mechanism of AD is not clear and there is no completely effective treatment for AD.Therefore,early diagnosis and prediction are crucial for the control of AD.In recent years,with the development of neural biomarkers for the recognition of AD,people’s understanding of the disease has changed from symptom based to brain based.Such changes occur before the onset of clinical symptoms,suggesting that AD as it has already been abbreviated before,which provides effective theoretical support for early diagnosis techniques.AD is increasingly recognized as a disorder of regional connectivity in the brain,but the asymmetry of the hemispheric network topology remains unclear in the preclinical phase of AD.In this study,Diffusion Tensor Imaging eterministic brain bundle angiography and the framework of standard graph theory will be adopted to systematically explore the hemispheric asymmetry of the brain structural network of patients in the pre-clinical stage for the first time,and to explore whether there is a correlation between abnormal hemispheric asymmetry and clinical variables,so as to provide an auxiliary reference for the early auxiliary diagnosis and prediction of AD.In recent years,there have been some researches on AD assisted diagnosis based on magnetic resonance imaging,but it is mainly based on functional brain network.However,the reliability of functional network constructed by different functional connection definition methods is different,so it cannot be used as a reliable standard.In order to ignore the structural features of networks in traditional studies,we propose a study based on the lateralization features of structural brain networks.The structural and functional asymmetry between the left and right hemispheres of the brain is called brain lateralization.At present,little is known about the structural network lateralization in AD continuous spectrum population,especially in the preclinical phase.Therefore,it is of great significance to the early diagnosis of AD if we can systematically study the topological changes of the abnormal structural brain network in AD pedigree population with the help of graph theory,and find out the differential characteristics that can effectively identify and distinguish the population at various stages,especially the early SCD population.This study mainly explored the changes of brain network attribute lateralization in the brain anatomical structure of the normal person(NC),subjective cognitive decline(SCD),amnestic mild cognitive impairment(a MCI),and AD in the four groups of subjects,and applied the lateralization feature in the diagnosis of the disease,providing a basis for the early auxiliary diagnosis of AD.The main contents of the study include:(1)construct a hemispherical structural brain network based on AAL brain template,and calculate the properties of structural brain network;(2)statistical analysis was conducted on the topological properties of hemispheric structural brain network,and the differences between left and right hemispheres in the structural brain network properties of AD spectrum population were analyzed,and the brain regions with significant differences were compared.(3)calculate the lateralization index of hemispheric structural network topology attribute,conduct statistical test on the asymmetric score,and explore the variation trend of the asymmetric score of structural brain network attribute in AD spectrum population;(4)statistical analysis was conducted on the clinical scale values of AD spectrum population,and correlation analysis was conducted between the lateralization index of structural network attributes and the clinical scale values.(5)use the SVM(support vector machine)classifier to classify the selected features,and analyze the difference in classification accuracy between the subjects with biased features and the subjects without biased features.In this study,we constructed a structural network of cerebral hemispheres and used graph theory to analyze the changes of structural brain network property lateralization in the population of alzheimer’s disease spectrum.Patients with NC to SCD,a MCI,and AD were found to have a gradual decrease in left-handedness.By analyzing the correlation between the scale and the network topology properties,it was found that the network topology asymmetry was related to the patient’s clinical characteristics and memory performance.Our results indicate that the abnormal hemispheric asymmetry in the network properties of the brain anatomical structure is related to the abnormal neurodegeneration of the AD spectrum population,and the inclusion of lateralization features can improve the classification accuracy of the subjects.This study reveals the underlying physiological and pathological mechanisms of human cognitive decline.It also provides a potential neural biomarker for the early diagnosis of AD,which is of great significance for the prevention of AD.