Intitulé de l'offre : Post-doctorate (M/W): “Online decoding of motor imagery using non-invasive measures of mu and beta burst brain activity” (H/F)
Référence : UMR5229-JAMBON-009
Nombre de Postes : 1
Lieu de travail : BRON
Date de publication : mercredi 15 novembre 2023
Type de contrat : CDD Scientifique
Durée du contrat : 13 mois
Date d'embauche prévue : 8 janvier 2024
Quotité de travail : Temps complet
Rémunération : from €2,934 gross monthly according to experience
Niveau d'études souhaité : Niveau 8 - (Doctorat)
Expérience souhaitée : Indifférent
Section(s) CN : Brain, cognition and behaviour
This position is available for a young researcher who has obtained a doctoral thesis in neuroimaging, neuroscience or computer science and signal processing with experience in processing human electrophysiology data, or even specifically in the field of brain-computer interfaces.
This position is open as part of the ANR HiFi project, under the co-supervision of Jérémie Mattout, PhD (co-leader of the COPHY team at the Lyon Neurosciences Research Center (CRNL)) and James Bonaiuto, PhD (head of the DANC team at the Marc Jeannerod Institute of Cognitive Sciences (ISCMJ)).
The mission associated with this position is to design, implement and then empirically validate a non-invasive brain-machine interface using EEG, and exploiting the original methods developed in our two teams, to extract transient (burst) activities, in the beta and mu frequency bands, in order to decode and translate into command, motor imagery, in different groups of participants.
The candidate will be involved in an ambitious and cutting-edge project aimed at developing real-time signal processing methods to efficiently, accurately and robustly translate sensorimotor activities measured with EEG during movement imagination tasks, for brain-machine interface applications. The objective of the project is to demonstrate that it is possible to exceed the performance of current methods, in particular by exploiting an important property of these activities, namely their occurrence in the form of transient bursts whose frequency of occurrence, form, amplitude and location in the cortex, encode important information.
The candidate will:
• Will have access to several public datasets or original data that she/he will contribute to acquire;
• Will develop and implement a set of methods for analyzing these data, initially by mimicking their processing in real time, to evaluate their performance;
• Will propose mathematical and algorithmic solutions to maximize this performance, by exploring different methodological avenues (Riemannian geometry, neural networks, etc.);
• Will conduct an experimental validation campaign of the developed approach;
• Will promote and disseminate this work through publications in international peer-reviewed journals and by participating in national and international conferences in the field;
• Will finally consider demonstrating the validity of this new approach, by participating in the future international competition “Cybathlon 2024”).
Enthusiastic and self-motivated candidates are encouraged to apply.
* Ph.D. in Neuroscience, Computer Science, Engineering, Machine Learning or related field
* Autonomy in programming in Python
* Experience in using imaging analysis software (MNE, SPM, Fieldtrip …)
* Experience in acquiring and analyzing EEG data
* Experience in implementing and experimenting non-invasive brain-computer interfaces
* Fluency and solid writing skills in English are expected
Contexte de travail
The successful candidate will benefit from structured mentoring support and opportunities for travel and presentation at national meetings. Both the CRNL and the ISCMJ are multidisciplinary research institutes offering a vibrant working atmosphere in an international environment with multiple opportunities for seminars, workshops and methodological clubs.
Contraintes et risques