Editorial: Functional Brain Mapping of Epilepsy Networks: Methods and Applications

• Published in: Frontiers in neuroscience Vol. 13; p. 417
• Main Authors: Abbott, David F., Archer, John S., Carney, Patrick W., Vaughan, David N., Jackson, Graeme D.
• Format: Journal Article
• Language: English
• Published: Switzerland Frontiers Research Foundation 21.05.2019; Frontiers Media S.A
• Subjects: Brain mapping; Brain research; Conflicts of interest; Convulsions & seizures; default mode network; Editorials; EEG; Electroencephalography; Epilepsy; fMRI; functional connectivity; Functional magnetic resonance imaging; Funding; interictal epileptiform discharge; Medical imaging; Mental health; Neuroimaging; Neuroscience; Neurosciences; NMR; Nuclear magnetic resonance; Australia; epilepsy; fMRI; seizure prediction; functional connectivity; EEG; default mode network; interictal epileptiform discharge; computational modeling and simulation
• ISSN: 1662-453X; 1662-4548; 1662-453X
• DOI: 10.3389/fnins.2019.00417

With technology such as simultaneous electroencephalography and functional magnetic resonance imaging (EEG-fMRI) now more readily available, it is possible to non-invasively map epileptiform activity throughout the entire brain at millimeter resolution. In application to MTLE and healthy controls, three distinct reliable networks were revealed: two that exhibited increased activity in patients (a network including hippocampus and amygdala bilaterally, and a network including postcentral gyri and temporal poles), and a network identified as specific to healthy controls (i.e., effectively decreased in patients, consisting of bilateral precuneus, anterior cingulate, thalamus, and parahippocampal gyrus). Confirming with computer modeling and phantom measurements that lead positioning can have a substantial effect on the amplitude of the MRI gradient artifact present on the EEG, they optimized the positions in a novel cap design. [...]whilst this substantially reduced gradient artifact amplitude on the phantom, it made things worse when used on human subjects. [...]improvements in model accuracy are required if one is to make accurate predictions for the human context.