Skip to Main content Skip to Navigation

Multimodal neurofeedback based on EEG/fMRI imaging techniques and visuo-haptic feedback for stroke rehabilitation

Mathis Fleury 1, 2, 3
2 Empenn
INSERM - Institut National de la Santé et de la Recherche Médicale, Inria Rennes – Bretagne Atlantique , IRISA-D5 - SIGNAUX ET IMAGES NUMÉRIQUES, ROBOTIQUE
3 Hybrid - 3D interaction with virtual environments using body and mind
Inria Rennes – Bretagne Atlantique , IRISA-D6 - MEDIA ET INTERACTIONS
Abstract : Neurofeedback (NF) is a technique that consists of sending back to an individual information on his brain activity, thus allowing him to modulate it. NF has thus been studied as a tool for brain rehabilitation in a large number of neurological and psychiatric disorders, and in particular for post-stroke rehabilitation. In this thesis, we have proposed and studied new multimodal NF systems, both at the input level, by combining multiple neuroimaging modalities - in particular Electroencephalography (EEG) and Functional Imaging Magnetic Resonance (fMRI), and at the output level, by proposing multimodal feedback combining visual and haptic feedback. In the first part of this thesis, we studied the possibility of combining visual and haptic feedback for a NF task. In a first step, we showed that the combination of visual feedback (3D moving hand), with the association of vibrations on the wrists, produced illusions of movement more intense than the use of a static hand or without feedback. In a second step, we showed that the use of visual-haptic (VH) feedback combined with a motor imaging (MI) task produced higher activations than with the MI task alone. Finally, we studied and implemented this VH feedback in the context of an MI-NF-IRMf study, where this feedback was confronted with the same feedback but visual alone (V) and haptic alone (H). Analysis of the NF scores and fMRI-fMRI activations suggests that this VH feedback led to more intense activations in the motor cortex than the H and V feedback alone and therefore may be potentially promising for stroke rehabilitation based on fMRI-NF. In the second part of this thesis, we have implemented an algorithm to locate the position of the EEG electrodes within EEG-fMRI experiments, which may prove useful in future EEG-fMRI-NF experiments. Finally, we present a multimodal, multi-session, EEG-fMRI-NF study with four stroke patients. Results suggest that two out of four patients benefited from NF training and reported significant functional gain, even though they were in the chronic phase of stroke.
Complete list of metadata
Contributor : Abes Star :  Contact
Submitted on : Wednesday, June 2, 2021 - 2:18:08 PM
Last modification on : Tuesday, October 19, 2021 - 11:04:41 AM


Version validated by the jury (STAR)


  • HAL Id : tel-03246556, version 2


Mathis Fleury. Multimodal neurofeedback based on EEG/fMRI imaging techniques and visuo-haptic feedback for stroke rehabilitation. Computer Vision and Pattern Recognition [cs.CV]. Université Rennes 1, 2021. English. ⟨NNT : 2021REN1S005⟩. ⟨tel-03246556v2⟩



Les métriques sont temporairement indisponibles