PhD position in Self-Supervised Learning for Anomaly Detection in Medical Neuroimaging

France Life Imaging; Villeurbanne, Grenoble; Sur place; EUR 40 000 - 60 000; Plein temps

Résumé du poste

Une entreprise de recherche dynamique recherche un candidat pour un doctorat orienté sur l'apprentissage auto-supervisé dans la détection d'anomalies en neuroimagerie. Ce projet, financé par le programme 'Défi IA', implique de travailler avec des experts en traitement d'images et en apprentissage automatique, cherchant à améliorer les performances des modèles via des méthodes innovantes. Le candidat idéal possède une formation en informatique et une passion pour la recherche scientifique.

Qualifications

• Étudiant(e) en doctorat avec expérience en apprentissage automatique.
• Compétences en traitement d'images et neuroimagerie.
• Capacité à travailler en équipe avec des chercheurs experts.

Responsabilités

• Développer des méthodes de détection et de segmentation auto-supervisées.
• Explorer des modèles d'apprentissage non supervisé.
• Collaborer avec les équipes de recherche de GIN et CREATIS.

Connaissances

Machine learning

Deep Learning

Neuroimaging

Segmentation

Self-supervised learning

Formation

Master en informatique ou domaine similaire

Description du poste

Type de structure : We offer a stimulating research environment gathering experts in Image processing, Neurosciences & Neuroimaging, Advanced Statistical and Machine Learning methods from CREATIS, Grenoble Institute of Neurosciences (GIN) and INRIA. The PhD position is granted by the “Défi IA” program sponsored by la Région Auvergne Rhône-Alpes.
The position is available in the framework of DAISIES project.

Contexte et mission : Scientific context

The vast majority of deep learning architectures for medical image analysis are based on supervised models requiring the
collection of large datasets of annotated examples. Building such annotated datasets, which requires skilled medical experts,
is time consuming and hardly achievable, especially for some specific tasks, including the detection of small and subtle lesions
that are sometimes impossible to visually detect and thus manually outline. This critical aspect significantly impairs
performances of supervised models and hampers their deployment in clinical neuroimaging applications, especially for brain
pathologies that require the detection of small size lesions (e.g. multiple sclerosis, microbleeds) or subtle structural or
morphological changes (e.g. Parkinson disease).

OBJECTIVE AND RESEARCH PROGRAM :

To solve this challenging issue, the objective of this thesis is to develop and evaluate deep self-supervised detection and
segmentation approaches whose training does not require any fine semantic annotations of the anomalies localization. We will
explore different categories of self-supervised methods, including: novel unsupervised auto-encoder based anomaly detection
models leveraging on the recent developments in visual transformers blocks (ViT) or vector quantized variational autoencoders
(VQ-VAE), scalability of Gaussian mixture models as well as weakly supervised models based on scarce annotations.

Key words: Machine learning, Deep Learning; Multidimensional data, Segmentation, Neuroimaging, Self-supervised learning,
Anomaly detection, Unsupervised representation learning

Starting date: Autumn 2022

How to apply: Send an email directly to three supervisors with your CV and persons to contact. Interviews of the selected
applicants will be done on an ongoing basis. Applications will be accepted up to the 30st of June.

Lieu : Location: Grenoble Neurosciences Institute: https://neurosciences.univ-grenoble-alpes.fr & CREATIS - Villeurbanne: https://www.creatis.insa-lyon.fr/. Time sharing in the two laboratories will be discussed with the selected candidates.

Contact : The PhD candidate will be co-supervised by: - GIN - team «Functional neuroimaging and brain perfusion»: Michel Dojat ([e-mail]), - CREATIS - team Myriad : Carole Lartizien ([e-mail]) - INRIA - team Statify: Florence Forbes ([e-mail])