Skip to main content
Aller à la page d’accueil de la Commission européenne (s’ouvre dans une nouvelle fenêtre)
français fr
CORDIS - Résultats de la recherche de l’UE
CORDIS

Reading the mind’s eye at 7 Tesla - A fMRI-based communication brain-computer interface for severely motor-impaired patients

Objectif

The Advanced ERC project ColumnarCodeCracking has pioneered ultra-high field fMRI at 7 Tesla for sub-millimeter neuroscience applications targeting cortical columns and cortical layers. One sub-project of the ERC project was to explore whether it is possible to build new brain computer interfaces (BCIs) exploiting the strong signal quality and higher resolution achievable at ultra-high magnetic fields. As part of our research (Emmerling et al., 2017, http://cordis.europa.eu/news/rcn/124885_en.html(s’ouvre dans une nouvelle fenêtre)) we discovered that it is possible to reconstruct letter shapes from activity in early visual areas that are merely imagined by participants during 7 Tesla fMRI scanning. Importantly, we could demonstrate that imagined letter shapes can be decoded from single imagination events of about 10 seconds without the need to average across multiple repetitions. These observations stimulated the idea for this PoC application, namely to use letter imagery for the first time as a communication BCI. We also have tested deep learning auto-encoder networks as part of the analysis and observed that these tools substantially increase the robustness of letter reconstruction. The three major goals of this PoC are 1) to perform 7 Tesla fMRI experiments with healthy participants to evaluate whether decoding brain activity patterns during letter imagery can be performed robust enough to be used as a communication BCI for severely motor-impaired (locked-in) patients, 2) to develop a BCI/neurofeedback software performing all required advanced online analyses, and 3) evaluate whether showing the online decoded letter during imagery helps participants to fine-tune the resulting shape. While 7 Tesla fMRI is not yet widely available in clinical settings, we aim to prepare first tests of the developed prototype with locked-in patients.

Champ scientifique (EuroSciVoc)

Programme(s)

Thème(s)

Régime de financement

ERC-POC - Proof of Concept Grant

Voir tous les projets financés dans le cadre de ce programme de financement

Appel à propositions

(s’ouvre dans une nouvelle fenêtre) ERC-2017-PoC

Voir tous les projets financés au titre de cet appel

Institution d’accueil

UNIVERSITEIT MAASTRICHT
Contribution nette de l'UE
€ 148 530,00
Adresse
MINDERBROEDERSBERG 4
6200 MD Maastricht
Pays-Bas
Région
Zuid-Nederland Limburg (NL) Zuid-Limburg
Type d’activité
Higher or Secondary Education Establishments
Liens
Coût total
€ 148 530,00

Bénéficiaires (1)