Vous êtes ici : Kiosque » Annonce

Identification

Identifiant: 
Mot de passe : 

Mot de passe oublié ?
Détails d'identification oubliés ?

Annonce

1 décembre 2023

(INT Marseille) Ingénieur de recherche en IA et analyse d'images IRM cérébrales


Catégorie : Ingénieur


Deep Learning for brain MR image reconstruction and analysis of large datasets

 

Duration: 2 years, can be extended to 4 years.

Location: Institut de Neurosciences de la Timone (INT), Marseille, France

Salary : [1900; 2500] euros/month after tax, depending on experience.

 

We are looking for a research engineer to perform reconstruction of MR brain fetal data and analysis of very large datasets of brain MRI, using deep learning techniques.

Full job description at https://url.univ-amu.fr/mlbrain

 

Context : The Institut de Neurosciences de la Timone (INT) is a large neuroscience institute at Aix-Marseille University that performs fundamental, translational, and clinical research at all scales, on various cognitive functions, and on various models, from rodents to human via different species of primates. INT owns a MRI acquisition facility that hosts a state-of-the-art 3T Siemens Prisma with the ability to scan healthy and pathological humans, as well as primates such as marmosets, macaques, and baboons. A large storage and computing infrastructure is available at INT for all data preprocessing and analysis.

Within INT, the MeCA research team is an interdisciplinary team that performs research on the organization and development of the cortex using MRI. MeCA has a long-lasting expertise in computational neuroanatomy, machine learning and image processing techniques applied to large MRI datasets. We implement our methods in open-source packages https://github.com/brain-slam/slam and software https://brainvisa.info/web/. We recently granted two projects on these topics, funded by the French National Research Agency (ANR).

 

Job definition: We are looking for a research engineer to join the MeCA team and be in charge of implementing specific image processing tools and their application to large datasets. More specifically, two projects are directly linked to this position.

The first one has to do with fetal MR brain imaging to study cortical development and its link with connectivity development. The applicant will be in charge of implementing an image processing pipeline combining several key steps: 1- the super-resolution reconstruction of anatomical images and diffusion-weighted (DWI) images, using machine learning-based algorithms [1-4]; 2-tissue segmentation based on recent deep-learning techniques like [5]; 3-extraction and modeling of the cortical surface; 4- diffusion model reconstruction for DWI data.

The second projects deal with the use of the UK Biobank database in the context of improving the fairness of machine-learning algorithms for analyzing large populations (up to 70k subjects). The applicant will be in charge of running existing in-house analysis and modeling pipelines to generate graph-based representation of cortical folding [6] on very large amount of data.

 

[1] Delannoy et al. (2020). Computers in Biology and Medicine, https://doi.org/10.1016/j.compbiomed.2020.103755

[2] Ebner et al. (2020). NeuroImage, https://doi.org/10.1016/j.neuroimage.2019.116324

[3] Deprez et al. (2019). IEEE Transactions on Medical Imaging. https://doi.org/10.1109/TMI.2019.2943565

[4] Khan et al. (2019). Neuroimage. https://doi.org/10.1016/j.neuroimage.2018.08.030

[5] Isensee et al. (2020). Nature Methods, http://dx.doi.org/10.1038/s41592-020-01008-z

[6] Takerkart et al. (2015) Medical Image Analysis, https://doi.org/10.1016/j.media.2016.04.011

 

Required skills :

Candidate must have a Master or PhD in data science and/or image analysis.

Proficiency in Python is necessary.

A good understanding of deep learning for medical imaging (segmentation, fine-tuning) is necessary.

Previous experience with MR imaging would be a plus but not mandatory.

 

How to apply:

Send CV, motivation letter, and two reference contact, to:

olivier.coulon@univ-amu.fr

guillaume.auzias@univ-amu.fr

sylvain.takerkart@univ-amu.fr

 

Dans cette rubrique

(c) GdR IASIS - CNRS - 2024.