Abstract
Brain networks are spatiotemporal phenomena that dynamically vary over time. Functional imaging approaches strive to noninvasively estimate these underlying processes. Here, we propose a novel source imaging approach that uses high-density EEG recordings to map brain networks. This approach objectively addresses the long-standing limitations of conventional source imaging techniques, namely, difficulty in objectively estimating the spatial extent, as well as the temporal evolution of underlying brain sources. We validate our approach by directly comparing source imaging results with the intracranial EEG (iEEG) findings and surgical resection outcomes in a cohort of 36 patients with focal epilepsy. To this end, we analyzed a total of 1,027 spikes and 86 seizures. We demonstrate the capability of our approach in imaging both the location and spatial extent of brain networks from noninvasive electrophysiological measurements, specifically for ictal and interictal brain networks. Our approach is a powerful tool for noninvasively investigating large-scale dynamic brain networks.
Highlights
Brain networks are spatiotemporal phenomena that dynamically vary over time
Various engineering approaches have been designed to improve the spatial resolution of these modalities, with the most promising advancements coming from a set of approaches called the electrophysiological source imaging (ESI) techniques[1,7,8,9,10]
Our method, was motivated by and tested in the framework of epilepsy imaging, we would like to emphasize that it has the capability of studying other normal and pathological brain functions, as underlying networks constituting these brain states are spatiotemporal processes that vary over relatively short time intervals[1]
Summary
Brain networks are spatiotemporal phenomena that dynamically vary over time. Functional imaging approaches strive to noninvasively estimate these underlying processes. Given brain’s organization, i.e. that spatially localized areas specialize in particular functions (functional segregation) and that the communication among these areas results in observed behavior (functional integration), the brain activity on large scales, can be modeled as a globally distributed spatiotemporal network[2]. Estimation of these spatiotemporally distributed networks, from noninvasive measurements, provides useful information about which brain areas are activated (or deactivated) during particular brain functions (normal or pathological), how extended these regions are, how their activity develops over time, and how multiple areas interact with each other during normal or pathological brain states. Different brain regions are functionally specialized to perform particular functions[2], determining the spatial extent of these regions is indispensable, and highly desirable when imaging brain networks
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
