A Research Of Cortical Dementia Based On Source-based Morphometry And Structural MRI

Part 1:Automatic brain structural features extraction and classification in Alzheimer’s disease based on independent component analysis and support vector machine.Purpose: By utilizing a data-driven method,we extract brain structural features which can reflect the variation among AD-MCI-NC populations,and using these features in classificationand compare its performance to brain atlases based methods.Method:We downloaded structural MRI images of AD(Alzheimer’s disease,151),MCI(mild cognitive impairment,124)and NC(normal control,171)populations from ADNI database.A source-based morphometry(SBM)analysis was performed on these data,which extract the components reflecting the variation of the voxel-based morphometry(VBM)results across subjects through independent component analysis(ICA).Besides,existing brain atlases(AAL90 and AAL 1024)was used to extract regional volumes as features.Support vector machine(SVM)was used to classify each pair among 3 populations with SBM components or atlas-based features.Age,gender and education years was controlled as covariances.We employed leave-one-out method for classification test.For each pair in AD,MCI and NC.Result:The SBM method was superior to the two atlas-based method on sensibility,specificity and overall accuracy.The area under curve(AUC)value of the receiver operating characteristic(ROC)curve of SBM method was significantly larger than those of AAL90 and AAL1024 method,indicating a better effect of classification.Conclusion:We employed a data-driven method for feature extraction,and observed better performance in classification than atlas-based method.These findings might be helpful for classification of AD,MCI and NC populations,which provide assistance to the early diagnosis and intervention of Alzheimer’s disease.Part 2:A causal structural covariance network research of brain atrophy pattern in Alzheimer’s disease.The structural covariance network(SCN)has provided a perspective on the large-scale brain organization impairment in the Alzheimer’s Disease(AD)continuum.However,the successive structural impairment across brain regions,which may underlie the disrupted SCN in the AD continuum,is not well understood.In the current study,we enrolled 446 subjects with AD,mild cognitive impairment(MCI)or normal ageing(NC)from the Alzheimer’s Disease Neuroimaging Initiative(ADNI)database.The SCN as well as a casual SCN(Ca SCN)based on Granger causality analysis were applied to the T1-weighted structural magnetic resonance images of the subjects.Compared with that of the NCs,the SCN was disrupted in the MCI and AD subjects,with the hippocampus and left middle temporal lobe being the most impaired nodes,which is in line with previous studies.In contrast,according to the 253 subjects with records on CSF amyloid and Tau,the Ca SCN revealed that during AD progression,the Ca SCN was enhanced.Specifically,the hippocampus,thalamus,and precuneus/posterior cingulate cortex(PCC)were identified as the core regions in which atrophy originated and could predict atrophy in other brain regions.Taken together,these findings provide a comprehensive view of brain atrophy in the AD continuum and the relationships among the brain atrophy in different regions,which may provide novel insight into the progression of AD.Part 3:Btrain structural features extraction and classification in Alzheimer’s disease and frontotemporal dementia based on independent component analysis and support vector machine.Purpose:To find differences among AD patients,FTD patients and healthy people and verify the validity of SVM method in identification of three groups.Method:We acquired structural MRI images of AD patients(69),FTD patients(69)and healthy people(72)from ADNI and NIFD database.Support vector machine was applied to classification research in three groups,with brain components set as signatures and age,gender,education years set as control variates.Half of the subjects were randomly selected as learn set and the rest as test set,and result of classification research was recorded.Result: The SVM method presented high accuracy in AD-FTD-NC classification research and the featural regions of brain atrophy in AD and FTD extracted were consistent with past researches.Conclusion:We employed a data-driven method and support vector machine in classification in AD-FTD-NC image data and observed high efficiency,verified the validity of SVM method in identifying dementia patients conhort with higher complexity.These findings are helpful to identify AD-FTD-NC patients,which provides assistance to early diagnosis and clinical intervention for dementia.