Predicting functional networks from region connectivity profiles in task-based versus resting-state fMRI data

Abstract

Intrinsic Connectivity Networks, patterns of correlated activity emerging from “resting-state” BOLD time series, are increasingly being associated with cognitive, clinical, and behavioral aspects, and compared with patterns of activity elicited by specific tasks. We study the reconfiguration of brain networks between task and resting-state conditions by a machine learning approach, to highlight the Intrinsic Connectivity Networks (ICNs) which are more affected by the change of network configurations in task vs. rest. To this end, we use a large cohort of publicly available data in both resting and task-based fMRI paradigms. By applying a battery of different supervised classifiers relying only on task-based measurements, we show that the highest accuracy to predict ICNs is reached with a simple neural network of one hidden layer. In addition, when testing the fitted model on resting state measurements, such architecture yields a performance close to 90% for areas connected to the task performed, which mainly involve the visual and sensorimotor cortex, whilst a relevant decrease of the performance is observed in the other ICNs. On one hand, our results confirm the correspondence of ICNs in both paradigms (task and resting) thus opening a window for future clinical applications to subjects whose participation in a required task cannot be guaranteed. On the other hand it is shown that brain areas not involved in the task display different connectivity patterns in the two paradigms.

Metrics for graph comparison: A practitioner’s guide

“…To further aggravate this situation, we note that the pattern of anomalous connectivity are isolated, while the vast majority of correlations are very close to zero (green cells in Fig 27). We conclude that the low amplitude of the contrastD i;j and its sparsity contribute to our inability to use graph distances to detect significant changes between the connectomes of the two populations (see also [117] for a detailed analysis of regional connectivity). We note that others have reported similar findings [118,119].…”

Spatiotemporal complexity patterns of resting‐state bioelectrical activity explain fluid intelligence: Sex matters

“…In the present study, we hypothesized a positive relationship between fluid intelligence and overall rsEEG complexity, especially in the FPN and its interactions. Since neural complexity is thought to reflect richness of brain signal (e.g., Garrett et al, 2013; Tononi et al, 1994) or its level of integrity (McIntosh et al, 2014; Sporns, Tononi, & Edelman, 2000), it is reasonable to believe that higher gf is associated with greater overall EEG signal complexity (Friston, 1996), especially in the resting‐state condition when brain activity is considered as a neural basis for specific tasks performance (Rasero et al, 2018). Existing evidence in that matter has yielded ambiguous outcomes.…”

Hubs disruption in mesial temporal lobe epilepsy. A resting‐state fMRI study on a language‐and‐memory network

“…Among the different FC measures that are available, FC at rest estimated from BOLD signals in fMRI is particularly robust for the description of the brain networks (van den Heuvel & Hulshoff Pol, ). Recent studies showed a very strong spatial similarity between intrinsic resting‐state networks and networks recruited by a variety of fMRI activation paradigms (Rasero et al, ). For instance, Cole et al found that cognitive task activations can be predicted in certain regions via estimated activity flow over resting‐state FC networks, for basic motor tasks but also for higher level tasks such as reasoning (Cole, Ito, Bassett, & Schultz, ).…”