Mise au point et définition du positionnement de nouveaux capteurs EEG compatibles et repérables en IRM : application à la localisation de source

Abstract : Spatial localization of scalp EEG electrodes is a major step in dipole source localization and it must be accurate, reproducible and practical. Several methods have been proposed in the last fifteen years. The most widely used method is currently electromagnetic digitization. In this work, we introduce a new automatic method for localizing and labeling EEG sensors using MRI (ALLES). First, we designed a new scalp EEG sensor which is MR compatible and localizable. Secondly, we validated this new technique on a head phantom and then in a clinical environment with normal volunteers and epileptic patients. To do this, we compare the reproducibility, accuracy and performances of our method with electromagnetic digitization. We also demonstrate that our method provides better reproducibility with a significant difference (p < 0.01). Concerning accuracy, both methods are equally accurate with no statistical differences. We have tested our method both on normal volunteers (SEP and VEP studies) and epileptic patients (Spikes studies). Source localizations were not influenced by ALLES and we observe results consistent with the literature. Finally, we develop a method which projects the surface positions of the sensors (10-10 system) onto the cortex. This tool is helpful for visual inspection of high resolution EEG traces and for electroclinical diagnostic. To conclude, automation makes our method (ALLES) very reproducible and easy to handle in a routine clinical environment. It offers the possibility of using MRI volume for both source localization and spatial localization of EEG sensors.
Complete list of metadatas

Cited literature [187 references]  Display  Hide  Download

https://hal.univ-lorraine.fr/tel-01777274
Contributor : Administrateur Du Ccsd <>
Submitted on : Tuesday, April 24, 2018 - 4:16:18 PM
Last modification on : Thursday, April 26, 2018 - 1:27:59 AM
Long-term archiving on : Wednesday, September 19, 2018 - 4:47:26 PM

File

SCD_T_2007_0150_KOESSLER.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : tel-01777274, version 1

Collections

Citation

Laurent Koessler. Mise au point et définition du positionnement de nouveaux capteurs EEG compatibles et repérables en IRM : application à la localisation de source. Médecine humaine et pathologie. Université Henri Poincaré - Nancy 1, 2007. Français. ⟨NNT : 2007NAN10150⟩. ⟨tel-01777274⟩

Share

Metrics

Record views

119

Files downloads

27