Brain network alterations in Alzheimer's disease measured by eigenvector centrality in fMRI are related to cognition and CSF biomarkers

Abstract

Recent imaging studies have demonstrated functional brain network changes in patients with Alzheimer's disease (AD). Eigenvector centrality (EC) is a graph analytical measure that identifies prominent regions in the brain network hierarchy and detects localized differences between patient populations. This study used voxel-wise EC mapping (ECM) to analyze individual whole-brain resting-state functional magnetic resonance imaging (MRI) scans in 39 AD patients (age 67 ± 8) and 43 healthy controls (age 69 ± 7). Between-group differences were assessed by a permutation-based method. Associations of EC with biomarkers for AD pathology in cerebrospinal fluid (CSF) and Mini Mental State Examination (MMSE) scores were assessed using Spearman correlation analysis. Decreased EC was found bilaterally in the occipital cortex in AD patients compared to controls. Regions of increased EC were identified in the anterior cingulate and paracingulate gyrus. Across groups, frontal and occipital EC changes were associated with pathological concentrations of CSF biomarkers and with cognition. In controls, decreased EC values in the occipital regions were related to lower MMSE scores. Our main finding is that ECM, a hypothesis-free and computationally efficient analysis method of functional MRI (fMRI) data, identifies changes in brain network organization in AD patients that are related to cognition and underlying AD pathology. The relation between AD-like EC changes and cognitive performance suggests that resting-state fMRI measured EC is a potential marker of disease severity for AD.

Keywords: Alzheimer's disease; amyloid-beta; cognition; functional connectivity; resting-state fMRI.

Figure 1

Differences between common centrality measures. a) Example of a simple graph. b) Degree centrality counts the number of edges attached to each vertex. c) Betweenness centrality counts how often a vertex occurs on the shortest path between two other vertices. d) Closeness centrality computes the average distance from a vertex to all other vertices, and differentiates between both central vertices and end vertices. e) EC counts the neighbors of each vertex, weighted by their centralities. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Figure 2

Mean EC maps for both groups. The voxel‐wise mean EC values were computed from all single subject maps in each group. For display purposes, data were resampled to a 1 × 1 × 1 voxel dimension. Red and yellow areas show high EC values, blue areas show low EC values. MNI coordinates: x = −1, y = −17, z = 37. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Figure 3

Group‐wise EC differences, after nonparametric permutation testing. Red voxels show clusters of EC increases in AD patients compared with controls, blue voxels show clusters of EC decreases in AD patients compared with controls (FWE‐corrected P < 0.05). Results are corrected for age and sex. These clusters were used as region‐of‐interest to extract mean EC values from each individual EC map. Results are displayed on standard MNI152 brain in radiological orientation. For display purposes, data were resampled to a 1 × 1 × 1 voxel dimension. MNI coordinates: x = −2, y = −84, z = 22. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Figure 4

Group‐wise gray matter density differences, displayed together with EC differences. Yellow voxels show regions of decreased gray matter density in AD compared with controls (FWE‐corrected P < 0.05). Red voxels show clusters of EC increases in AD patients compared with controls; voxels are green where EC increases overlap with gray matter density decreases in AD. Blue voxels show clusters of EC decreases in AD patients compared with controls; voxels are green where EC decreases overlap with gray matter density decreases in AD. Results are corrected for age and sex. Results are displayed on standard MNI152 brain in radiological orientation. For display purposes, data were resampled to a 1 × 1 × 1 voxel dimension. MNI coordinates: x = 5, y = −84, z = 19. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Figure 5

Scatterplots of the relationship between a) frontal EC increases and MMSE‐scores, b) occipital EC decreases and MMSE‐scores, c) frontal EC increases and amyloid in CSF, and d) occipital EC decreases and amyloid in CSF.