Payoux, P. | bioRxiv Science

bioRxiv2021

The European Prevention of Alzheimer's Dementia (EPAD) MRI Dataset and Processing Workflow

The European Prevention of Alzheimer Dementia (EPAD) is a multi-center study that aims to characterize the preclinical and prodromal stages of Alzheimers Disease. The EPAD imaging dataset includes core (3D T1w, 3D FLAIR) and advanced (ASL, diffusion MRI, and resting-state fMRI) MRI sequences. Here, we give an overview of the semi-automatic multimodal and multisite pipeline that we developed to curate, preprocess, quality control (QC), and compute image-derived phenotypes (IDPs) from the EPAD MRI dataset. This pipeline harmonizes DICOM data structure across sites and performs standardized MRI preprocessing steps. A semi-automated MRI QC procedure was implemented to visualize and flag MRI images next to site-specific distributions of QC features -- i.e. metrics that represent image quality. The value of each of these QC features was evaluated through comparison with visual assessment and step-wise parameter selection based on logistic regression. IDPs were computed from 5 different MRI modalities and their sanity and potential clinical relevance were ascertained by assessing their relationship with biological markers of aging and dementia. The EPAD v1500.0 data release encompassed core structural scans from 1356 participants 842 fMRI, 831 dMRI, and 858 ASL scans. From 1356 3D T1w images, we identified 17 images with poor quality and 61 with moderate quality. Five QC features -- Signal to Noise Ratio (SNR), Contrast to Noise Ratio (CNR), Coefficient of Joint Variation (CJV), Foreground-Background energy Ratio (FBER), and Image Quality Rate (IQR) -- were selected as the most informative on image quality by comparison with visual assessment. The multimodal IDPs showed greater impairment in associations with age and dementia biomarkers, demonstrating the potential of the dataset for future clinical analyses.

neuroscience↗

bioRxiv2020

Modular slowing of resting-state dynamic Functional Connectivity as a marker of cognitive dysfunction induced by sleep deprivation

Dynamic Functional Connectivity (dFC) in the resting state (rs) is considered as a correlate of cognitive processing. Describing dFC as a flow across morphing connectivity configurations, our notion of dFC speed quantifies the rate at which FC networks evolve in time. Here we probe the hypothesis that variations of rs dFC speed and cognitive performance are selectively interrelated within specific functional subnetworks. In particular, we focus on Sleep Deprivation (SD) as a reversible model of cognitive dysfunction. We found that whole-brain level (global) dFC speed significantly slows down after 24h of SD. However, the reduction in global dFC speed does not correlate with variations of cognitive performance in individual tasks, which are subtle and highly heterogeneous. On the contrary, we found strong correlations between performance variations in individual tasks -including Rapid Visual Processing (RVP, assessing sustained visual attention)- and dFC speed quantified at the level of functional subnetworks of interest. Providing a compromise between classic static FC (no time) and global dFC (no space), modular dFC speed analyses allow quantifying a different speed of dFC reconfiguration independently for sub-networks overseeing different tasks. Importantly, we found that RVP performance robustly correlates with the modular dFC speed of a characteristic frontoparietal module. HighlightsO_LISleep Deprivation (SD) slows down the random walk in FC space implemented by Dynamic Functional Connectivity (dFC) at rest. C_LIO_LIWhole-brain level slowing of dFC speed does not selectively correlate with fine and taskspecific changes in performance C_LIO_LIWe quantify dFC speed separately for different link-based modules coordinated by distinct regional "meta-hubs" C_LIO_LIModular dFC speed variations capture subtle and task-specific variations of cognitive performance induced by SD. C_LI Author summaryWe interpreted dynamic Functional Connectivity (dFC) as a random walk in the space of possible FC networks performed with a quantifiable "speed". Here, we analyze a fMRI dataset in which subjects are scanned and cognitively tested both before and after Sleep Deprivation (SD), used as a reversible model of cognitive dysfunction. While global dFC speed slows down after a sleepless night, it is not a sufficiently sensitive metric to correlate with fine and specific cognitive performance changes. To boost the capacity of dFC speed analyses to account for fine and specific cognitive decline, we introduce the notion of modular dFC speed. Capitalizing on an edge-centric measure of functional connectivity, which we call Meta-Connectivity, we isolate subgraphs of FC describing relatively independent random walks (dFC modules) and controlled by distinct "puppet masters" (meta-hubs). We then find that variations of the random walk speed of distinct dFC modules now selectively correlate with SD-induced variations of performance in the different tasks. This is in agreement with the fact that different subsystems - distributed but functionally distinct- oversee different tasks. The high sensitivity of modular dFC analyses bear promise of future applications to the early detection and longitudinal characterization of pathologies such as Alzheimers disease.

neuroscience↗