Abstract
Quantitatively assessing the medial temporal lobe (MTL) structures atrophy is vital for early diagnosis of Alzheimer’s disease (AD) and accurately tracking of the disease progression. Morphometry characteristics such as gray matter volume (GMV) and cortical thickness have been proved to be valuable measurements of brain atrophy. In this study, we proposed a morphometric MRI analysis based method to explore the cross-sectional differences and longitudinal changes of GMV and cortical thickness in patients with AD, MCI (mild cognitive impairment) and the normal elderly. High resolution 3D MRI data was obtained from ADNI database. SPM8 plus DARTEL was carried out for data preprocessing. Two kinds of z-score map were calculated to, respectively, reflect the GMV and cortical thickness decline compared with age-matched normal control database. A volume of interest (VOI) covering MTL structures was defined by group comparison. Within this VOI, GMV, and cortical thickness decline indicators were, respectively, defined as the mean of the negative z-scores and the sum of the normalized negative z-scores of the corresponding z-score map. Kruskal–Wallis test was applied to statistically identify group wise differences of the indicators. Support vector machines (SVM) based prediction was performed with a leave-one-out cross-validation design to evaluate the predictive accuracies of the indicators. Linear least squares estimation was utilized to assess the changing rate of the indicators for the three groups. Cross-sectional comparison of the baseline decline indicators revealed that the GMV and cortical thickness decline were more serious from NC, MCI to AD, with statistic significance. Using a multi-region based SVM model with the two indicators, the discrimination accuracy between AD and NC, MCI and NC, AD and MCI was 92.7, 91.7, and 78.4%, respectively. For three-way prediction, the accuracy was 74.6%. Furthermore, the proposed two indicators could also identify the atrophy rate differences among the three groups in longitudinal analysis. The proposed method could serve as an automatic and time-sparing approach for early diagnosis and tracking the progression of AD.
Highlights
Alzheimer’s disease (AD) is an insidious onset neurodegenerative disease primarily characterized by progressive memory loss and accompanied by several kinds of cognitive and functional impairment (McKhann et al, 2011)
It suggests that the atrophy pattern characterized by the combination of gray matter volume (GMV) and cortical thickness of Medial temporal lobe (MTL) structures may be useful to identify AD and Mild cognitive impairment (MCI) whose underlying pathophysiology is AD, and may overcome the specificity lacking for differentiating AD and MCI from other non-AD forms of dementia (Laakso et al, 1996; Chan et al, 2001; van de Pol et al, 2006; Bastos-Leite et al, 2007)
In this study, based on the Morphometric Analysis Program (MAP) framework, we proposed a modified morphometric MRI analysis method to quantitatively assessed the GMV and cortical thickness decline in MTL structures of AD, MCI, and normal control (NC), and to validate the hypothesis that the degrees and rates of atrophy in MTL of AD, MCI and normal aging are different, from severe to slight
Summary
Alzheimer’s disease (AD) is an insidious onset neurodegenerative disease primarily characterized by progressive memory loss and accompanied by several kinds of cognitive and functional impairment (McKhann et al, 2011). The pattern of volume reduction of the hippocampal subfields combined with the cortical thinning of the adjacent extrahippocampal structures such as entorhinal and perirhinal cortex and parahippocampal cortex was found specific for AD compared with dementia with Lewy bodies (Delli Pizzi et al, 2016; Mak et al, 2016; Pettigrew et al, 2016) It suggests that the atrophy pattern characterized by the combination of GMV and cortical thickness of MTL structures may be useful to identify AD and MCI whose underlying pathophysiology is AD, and may overcome the specificity lacking for differentiating AD and MCI from other non-AD forms of dementia (Laakso et al, 1996; Chan et al, 2001; van de Pol et al, 2006; Bastos-Leite et al, 2007)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
