Accès direct au contenu

x

Moteur de recherche interne

Grenoble Institut des Neurosciences Grenoble Institut des Neurosciences

  • Youtube
  • Linkedin
  • Twitter

Accueil > Nous rejoindre > Offres d'emploi

Accéder au plan complet du site

Post-Doctoral position

Background: The general aim of the ‘MR FUSE’ project, funded by the ANR, is the development of new Magnetic Resonance (MR) imaging tools for the management of acute stroke patients. We propose to use the concept of MR ‘fingerprinting’ (ref1) combined with artificial intelligence algorithms to create fast (<6 min), efficient (6 parameters in one acquisition) and robust (motion insensitive) MRI protocols that don’t require the use of contrast agent. These imaging solutions will allow a better triage of acute stroke patients to appropriate treatments. MR FUSE will also give us access to new biomarkers (microvascular network integrity, hypoxia, etc…) (ref2) that will be used to improve the evaluation of the ischemic penumbra and lead to better individualization of treatments outside of the current therapeutic window. MR FUSE will be optimized with numerical simulations and deep learning tools, validated in a preclinical study and tested in healthy volunteers to prepare for a future clinical trial.

Goal: The proposed work combines physics and applied mathematics (mostly AI). It consists in developing MRF tools for our preclinical and clinical platforms. The objectives are:

  • To develop MRF sequences for the measurements of standard MR parameters such as T1, T2, T2* or ADC. MRF developments don’t necessitate knowledge of MR vendors programming languages.
  • To use in-house tools to simulate MR signals coming from realistic voxels and to use deep learning networks to create dictionaries of numerical signals.
  • To collaborate with a PhD student on the development of AI codes that optimize MRF acquisition patterns (RF pulses, repetition times, phases…). The best patterns will be ‘fused’ to obtain a unique sequence.
  • To acquire in vivo data (human + small animal) and to test the tools when the level of inspired oxygenation is increased (hyperoxia) or decreased (hypoxia).

MRI simulations and AI codes will be written in Python (+keras, +PyTorch) and Matlab. Our multidisciplinary team belongs to the MIAI Grenoble Institute (https://miai.univ-grenoble-alpes.fr/). Data will be collected on Bruker et Philips scanners with the help of engineers from the IRMaGe platform (https://irmage.univ-grenoble-alpes.fr/).

Formation: PhD in Physics / Info / Applied Math. Experience in medical imaging and/or AI is preferred.

Contact: GIN / Team "Functional Neuroimaging and Brain Perfusion": Thomas Christen (thomas.christen@univ-grenoble-alpes.fr).

Location: Grenoble Institute of Neurosciences (GIN): https://neurosciences.univ-grenoble-alpes.fr/en/

Start: Winter 2021. Expected project duration: 24 months.

References: [ref1] Ma et al., Nature, 2013. [ref2] Christen et al., Neuroimage, 2014.


Téléchargement(s)

Mise à jour le 24 septembre 2021

Contacts

Pour les stages (master, licence, 3ème), envoyer directement un email au responsable de l'équipe que vous avez identifiée.

Pour une candidature spontanée pour un emploi et uniquement pour cela, envoyez un email à gincomm[at]univ-grenoble-alpes.fr ou utilisez le formulaire de contact.

Membres
Associés renforcés
Associés simples