Interrelating differences in structural and functional connectivity in the older adult's brain

Abstract

In the normal aging process, the functional connectome restructures and shows a shift from more segregated to more integrated brain networks, which manifests itself in highly different cognitive performances in older adults. Underpinnings of this reorganization are not fully understood, but may be related to age-related differences in structural connectivity, the underlying scaffold for information exchange between regions. The structure-function relationship might be a promising factor to understand the neurobiological sources of interindividual cognitive variability, but remain unclear in older adults. Here, we used diffusion weighted and resting-state functional magnetic resonance imaging as well as cognitive performance data of 573 older subjects from the 1000BRAINS cohort (55-85 years, 287 males) and performed a partial least square regression on 400 regional functional and structural connectivity (FC and SC, respectively) estimates comprising seven resting-state networks. Our aim was to identify FC and SC patterns that are, together with cognitive performance, characteristic of the older adults aging process. Results revealed three different aging profiles prevalent in older adults. FC was found to behave differently depending on the severity of age-related SC deteriorations. A functionally highly interconnected system is associated with a structural connectome that shows only minor age-related decreases. Because this connectivity profile was associated with the most severe age-related cognitive decline, a more interconnected FC system in older adults points to a process of dedifferentiation. Thus, functional network integration appears to increase primarily when SC begins to decline, but this does not appear to mitigate the decline in cognitive performance.

Prediction of cognitive performance differences in older age from multimodal neuroimaging data

“…For functional data, each matrix entry reflected the Pearson’s correlation of the average time series of two nodes. As an additional step, a statistical significant test of each correlation coefficient was performed making use of the Fourier transform and permutation testing (1000 repeats) to reduce the amount of spurious correlations [ 11 , 12 , 51 ]. Non-significant edges at p ≥ 0.05 were set to zero.…”

“…1 A). For both RSFC and SC, the focus was with nodal-level (1) within-network connectivity (400 features) defined as the sum of weights of one node attached to all nodes within its respective network divided by the total number of edges in the network, (2) inter-network connectivity (400 features) defined as the sum of weights from one node to all nodes outside its respective network divided by the number of edges in the network as well as (3) a ratio score (400 features) defined as within-network connectivity of a node in relation to its inter-network connectivity [ 12 ]. The total feature vector for each participant encompassed 2,800 features (1200 RSFC estimates + 1200 SC estimates + 400 region-wise GMV values).…”

“…To ensure data quality, mean GMV values were calculated and participants with values greater than 1.5 times the inter-quartile range were excluded from further analyses. Functional and diffusion tensor images were preprocessed according to an established pipeline by [12]. For all functional images, this included (1) deletion of the first four EPI volumes, (2) head movement correction using a two-pass procedure, (2) application of ICA-based Automatic Removal of Motion Artifacts (ICA-AROMA) [45] combined with global signal regression, (3) application of a band-pass filter (0.01-0.1 Hz), and (4) registration to MNI152 template using a unified segmentation approach [46].…”

“…The aging population experiences declines in many cognitive functions, e.g., memory and executive functions [1,2]. In groups of healthy older adults, agerelated cognitive decline has been partly explained by alterations in network architecture, structural (SC) and resting-state functional connectivity (RSFC) of major resting-state networks (RSNs), and grey matter (GM) atrophy [1,[3][4][5][6][7][8][9][10][11][12][13]. However, despite robust findings at the group level, cognitive performance has been found to vary greatly at the individual level [1,14], particularly in the older ages.…”

Probabilistic Inference on Virtual Brain Models of Disorders

“…During aging or some diseases, the SC is reported to deteriorate (Antonenko and Flöel, 2013;Damoiseaux, 2017;Zuo et al, 2017), or be perturbed (Stam, 2014;Ozdemir et al, 2020;Menardi et al, 2021), particularly with respect to the number of inter-and intra-hemispheric fibers within tracts and fiber density (Puxeddu et al, 2020;Petkoski et al, 2023;Lavanga et al, 2023). These observations, however, do not map trivially on functional data, such as fMRI (Stumme et al, 2022;Jockwitz et al, 2023;Krämer et al, 2023). This could be due to several factors, notably those related to the structure of the generative model family: these features include a degenerate mapping between Structural Connectivity (SC) and Functional Connectivity (FC), high dimensionality of the parameter space, and degeneracy induced by network effects in the latent state space, all introducing potential non-identifiability.…”