Combining electroencephalography and functional magnetic resonance imaging for neurofeedback

Lorraine Perronnet 1, 2
2 VisAGeS - Vision, Action et Gestion d'informations en Santé
INSERM - Institut National de la Santé et de la Recherche Médicale : U1228, Inria Rennes – Bretagne Atlantique , IRISA-D5 - SIGNAUX ET IMAGES NUMÉRIQUES, ROBOTIQUE
Abstract : NF is the process of feeding back real-time information to an individual about his/her ongoing brain activity, so that he/she can train to self-regulate neural substrates of specific behavioral functions. NF has been extensively studied for brain rehabilitation of patients with psychiatric and neurological disorders. However its effective deployment in the clinical armamentarium is being held back by the lack of evidence about its efficacy. One of the possible reason for the debated efficacy of current approaches could be the inherent limitations of single imaging modalities. Indeed, most NF approaches rely on the use of a single modality, EEG and fMRI being the two most widely used. While EEG is inexpensive and benefits from a high temporal resolution (millisecond), its spatial resolution (centimeters) is limited by volume conduction of the head and the number of electrodes. Also source localization from EEG is inaccurate because of the ill-posed inverse problem. In a complementary way, fMRI gives access to the self-regulation of specific brain regions at high spatial resolution (millimeter) but has low temporal resolution (second). Combined EEG-fMRI has proven much valuable for the study of human brain function, however it has rarely been exploited for NF purpose. In the context of NF, combining EEG and fMRI enables cross-modal paradigm evaluation and validation. But more interestingly it opens up avenues for the development of new NF approaches that would mix both modalities, either at the calibration phase or to provide a bimodal NF signal. Combined EEG-fMRI poses numerous challenges with regard to basic physiology, study design, data quality, analysis/integration and interpretation. These challenges are even greater if EEG and fMRI are both to be used simultaneously for online NF computation, because of the real-time constraint and the difficulty to find a task design compatible with EEG and fMRI' diverging natures. The theoretical part of this PhD dissertation aims at identifying methodological aspects that differ between EEG-NF and fMRI-NF and at examining the motivations and strategies for combining EEG and fMRI for NF purpose. Among these combination strategies, we choose to focus on bimodal EEG-fMRI-NF as it seems to be one of the most promising approach and is mostly unexplored. The feasibility of this approach was recently demonstrated and opened an entire new field of investigation. First and foremost, we would like to address the following questions: what is the added value of bimodal NF over unimodal NF; are there any specific mechanisms involved when learning to control two NF signals simultaneously; how to integrate EEG and fMRI to derive a single feedback ? The experimental part of this PhD dissertation therefore focuses on the development and evaluation of methods for bimodal EEG-fMRI-NF. In order to conduct bimodal NF experiments, we start by building up a real-time EEG-fMRI platform. Then in a first study, we compare for the first time bimodal EEG-fMRI-NF with unimodal EEG-NF and fMRI-NF. Eventually, in a second study, we introduce and evaluate two integrated feedback strategies for EEG-fMRI-NF.
Document type :
Theses
Complete list of metadatas

Cited literature [209 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-01661583
Contributor : Abes Star <>
Submitted on : Tuesday, December 12, 2017 - 1:19:11 AM
Last modification on : Friday, September 13, 2019 - 9:50:02 AM

File

PERRONNET_Lorraine.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-01661583, version 1

Citation

Lorraine Perronnet. Combining electroencephalography and functional magnetic resonance imaging for neurofeedback. Medical Imaging. Université Rennes 1, 2017. English. ⟨NNT : 2017REN1S043⟩. ⟨tel-01661583⟩

Share

Metrics

Record views

548

Files downloads

328