The Backbone Network of Dynamic FunctionalConnectivity

Temporal networks have become increasingly pervasive in many real-world applications, including the functional connectivity analysis of spatiallyseparated regions of the brain. A major challenge in analysis of such networksis the identification of noise confounds, which introduce temporal ties thatare non-essential, or links that are formed by chance due to local propertiesof the nodes. Several approaches have been suggested in the past for staticnetworks or temporal networks with binary weights for extracting significantties whose likelihood cannot be reduced to the local properties of the nodes.In this work, we propose a data-driven procedure to reveal the irreducible tiesin dynamic functional connectivity of resting state fRMI data with continu-ous weights. This framework includes a null model that estimates the latentcharacteristics of the distributions of temporal links through optimization,followed by a statistical test to filter the links whose formation can be reducedto the activities and local properties of their interacting nodes. We demon-strate the benefits of this approach by applying it to a resting state fMRIdataset, and provide further discussion on various aspects and advantages ofit.