Dataset Information

Voxel-based morphometry and a deep learning model for the diagnosis of early Alzheimer's disease based on cerebral gray matter changes.

ABSTRACT: This study aimed to analyse cerebral grey matter changes in mild cognitive impairment (MCI) using voxel-based morphometry and to diagnose early Alzheimer's disease using deep learning methods based on convolutional neural networks (CNNs) evaluating these changes. Participants (111 MCI, 73 normal cognition) underwent 3-T structural magnetic resonance imaging. The obtained images were assessed using voxel-based morphometry, including extraction of cerebral grey matter, analyses of statistical differences, and correlation analyses between cerebral grey matter and clinical cognitive scores in MCI. The CNN-based deep learning method was used to extract features of cerebral grey matter images. Compared to subjects with normal cognition, participants with MCI had grey matter atrophy mainly in the entorhinal cortex, frontal cortex, and bilateral frontotemporal lobes (p < 0.0001). This atrophy was significantly correlated with the decline in cognitive scores (p < 0.01). The accuracy, sensitivity, and specificity of the CNN model for identifying participants with MCI were 80.9%, 88.9%, and 75%, respectively. The area under the curve of the model was 0.891. These findings demonstrate that research based on brain morphology can provide an effective way for the clinical, non-invasive, objective evaluation and identification of early Alzheimer's disease.

SUBMITTER: Huang H

PROVIDER: S-EPMC9890469 | biostudies-literature | 2023 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Voxel-based morphometry and a deep learning model for the diagnosis of early Alzheimer's disease based on cerebral gray matter changes.

Huang Huaidong H Zheng Shiqiang S Yang Zhongxian Z Wu Yi Y Li Yan Y Qiu Jinming J Cheng Yan Y Lin Panpan P Lin Yan Y Guan Jitian J Mikulis David John DJ Zhou Teng T Wu Renhua R

Cerebral cortex (New York, N.Y. : 1991) 20230101 3

This study aimed to analyse cerebral grey matter changes in mild cognitive impairment (MCI) using voxel-based morphometry and to diagnose early Alzheimer's disease using deep learning methods based on convolutional neural networks (CNNs) evaluating these changes. Participants (111 MCI, 73 normal cognition) underwent 3-T structural magnetic resonance imaging. The obtained images were assessed using voxel-based morphometry, including extraction of cerebral grey matter, analyses of statistical diff ...[more]

PMID: 35301516

Similar Datasets

Project description:PurposeIncreasing evidence suggests that retinal microvasculature may reflect global cerebral atrophy. However, little is known about the relation of retinal microvasculature with specific brain regions and brain networks. Therefore, we aimed to unravel the association of retinal microvasculature with gray matter changes and structural covariance network using a voxel-based morphometry (VBM) analysis.MethodsOne hundred and forty-four volunteers without previously known neurological diseases were recruited from West China Hospital, Sichuan University between April 1, 2021, and December 31, 2021. Retinal microvasculature of superficial vascular plexus (SVP), intermediate capillary plexus (ICP), and deep capillary plexus (DCP) were measured by optical coherence tomography angiography using an automatic segmentation. The VBM and structural covariance network analyses were applied to process brain magnetic resonance imaging (MRI) images. The associations of retinal microvasculature with voxel-wise gray matter volumes and structural covariance network were assessed by linear regression models.ResultsIn the study, 137 participants (mean age = 59.72 years, 37.2% men) were included for the final analysis. Reduced perfusion in SVP was significantly associated with reduced voxel-wise gray matter volumes of the brain regions including the insula, putamen, occipital, frontal, and temporal lobes, all of which were located in the anterior part of the brain supplied by internal carotid artery, except the occipital lobe. In addition, these regions were also involved in visual processing and cognitive impairment (such as left inferior occipital gyrus, left lingual gyrus, and right parahippocampal gyrus). In regard to the structural covariance, the perfusions in SVP were positively related to the structural covariance of the left lingual gyrus seed with the left middle occipital gyrus, the right middle occipital gyrus, and the left middle frontal gyrus.ConclusionsPoor perfusion in SVP was correlated with reduced voxel-wise gray matter volumes and structural covariance networks in regions related to visual processing and cognitive impairment. It suggests that retinal microvasculature may offer a window to identify aging related cerebral alterations.

Project description:Results from studies that have examined age-related changes in gray matter based on structural MRI scans have not always been consistent. Reasons for this variability likely include small or unevenly-distributed samples, different methods for tissue class segmentation and spatial normalization, and the use of different statistical models. Particularly relevant to the latter is the method of adjusting for global (total) gray matter when making inferences about regionally-specific changes. In the current study, we use voxel-based morphometry (VBM) to explore the impact of these methodological choices in assessing age-related changes in gray matter volume in a sample of 420 adults evenly distributed between the ages of 18-77 years. At a broad level, we replicate previous findings, showing age-related gray matter decline in nearly all parts of the brain, with particularly rapid decline in inferior regions of frontal cortex (e.g., insula and left inferior frontal gyrus) and the central sulcus. Segmentation was improved by increasing the number of tissue classes and using less age-biased templates, and registration was improved by using a diffeomorphic flow-based algorithm (DARTEL) rather than a "constrained warp" approach. Importantly, different approaches to adjusting for global effects--not adjusting, Local Covariation, Global Scaling, and Local Scaling--significantly affected regionally-specific estimates of age-related decline, as demonstrated by ranking age effects across anatomical ROIs. Split-half cross-validation showed that, on average, Local Covariation explained a greater proportion of age-related variance across these ROIs than did Global Scaling. Nonetheless, the appropriate choice for global adjustment depends on one's assumptions and specific research questions. More generally, these results emphasize the importance of being explicit about the assumptions underlying key methodological choices made in VBM analyses and the inferences that follow.

Dataset Information

Voxel-based morphometry and a deep learning model for the diagnosis of early Alzheimer's disease based on cerebral gray matter changes.

Publications

Voxel-based morphometry and a deep learning model for the diagnosis of early Alzheimer's disease based on cerebral gray matter changes.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets