Meeting Banner
Abstract #3647

Influence of Volume Conductor Model Errors on EEG Dipole Source Localization in Neonates

Ivana Despotovic1, Perumpillichira J. Cherian2, Maarten De Vos3, Hans Hallez4, Paul Govaert5, Maarten Lequin6, Gerhard H. Visser2, Renate Renate M. Swarte5, Ewout Vans

1Ghent University, MEDISIP-IPI-IBBT, Ghent, Belgium; 2Erasmus MC-Sophia, University Medical Center Rotterdam, Department of Clinical Neurophysiology, Rotterdam, Netherlands; 3Katholieke Universiteit Leuven, ESAT-IBBT-K.U. Leuven Future Health Department, Leuven, Belgium; 4Ghent University, MEDISIP-IPI-IBBT, Ghent, Belgium; 5Erasmus MC-Sophia, Department of Neonatology, Rotterdam, Netherlands; 6Erasmus MC-Sophia, Department of Pediatric Radiology, Rotterdam, Netherlands

Integrating MRI and EEG data for 3D localization of active sources in the brain is an important diagnostic tool in patients with neurological disorders. However, this method has not yet been sufficiently studied in neonates. To investigate its feasibility, we developed an integrated method for dipole source localization in neonates based on a realistic head model. Here, we present our method and explore its sensitivity to electrode mislocalization and the variations in neonatal skull conductivity and geometry. Experimental results indicate that EEG source imaging is feasible in neonates and with further developments this technique can be a useful diagnostic tool.

Abstract PDF E-Poster Video

Keywords

adults almost anatomy animals anterior apply assigned assigning automatic available baker believe better bigger biological biology biomedical brain chosen closer clustering compartments components conductance conductivities conductivity conductor consists constrained construct correspond cubic deals decomposition despite detection developed development diagnostic dielectric dipole dipoles direct disorders distance electrodes electroencephalography engineer epileptic error errors estimation euclidean examine experiment experimental feasibility feasible fifth finally finite fitted fluid fontanel forward fourth fused fuzzy grid head health help hole important infants influence insufficient integrated integrating inter international inverse investigate knowledge lesions localization located location made material measured medical medicine might model modeled modeling models negligible neonatal neonates neurophysiology newborn next noticed omitting original outline part pediatric physicians physiologist place position problem propagation properties radiology reach realistic reasons reciprocity reconstruction resolution respectively scalp segmentation segmented seizure since skull slice source sources spatially studies table technical thinner third tissue tools types understand useful validation vicinity volume widely wrong wrongly