Eloi Navet 🧠
Eloi Navet

PhD Candidate in Computer Science (AI & Medical Imaging)

LaBRI - In2Brain, University of Bordeaux

About me

I am Eloi Navet, a computer science PhD candidate at the University of Bordeaux within the LaBRI and the In2Brain consortium. Since October 2024 I have been pursuing the doctoral project “Une Assemblée d’IA pour la Prédiction des Maladies Neurologiques”, registered with the Mathematics and Computer Science Doctoral School (EDMI). I design ensembles of deep learning models for 3D brain MRI to support differential diagnosis of Alzheimer’s disease, frontotemporal dementia, and dementia with Lewy bodies.

The project is supervised by Pierrick Coupé (LaBRI / CNRS), Rémi Giraud (IMS Bordeaux INP), and Boris Mansencal (LaBRI / CNRS) in collaboration with Bordeaux University Hospital and the Institute of Neurodegenerative Diseases (IMN).

Download CV
Interests
  • Artificial Intelligence
  • Applied Mathematics
  • Medical Imaging
  • Data Science
  • Remote Sensing
  • Bioinformatics
  • Open Science & Reproducibility
Education

  • Doctorate in Computer Science (ongoing)

    University of Bordeaux

  • Engineering Degree in Computer Science

    Grenoble INP ENSIMAG

  • Master of Applied Mathematics

    Université Grenoble Alpes

  • Exchange in semester Master of Science

    Aalto University

📚 My PhD Research

Thesis topic:
“Une Assemblée d’IA pour la Prédiction des Maladies Neurologiques”, a computer science doctorate at the University of Bordeaux.

Laboratory & team:
Research carried out at (UMR 5800, Images and Sound team) within the consortium since October 2024.

Supervision:
(thesis advisor, LaBRI / CNRS), (co-advisor, IMS Bordeaux INP), and (LaBRI / CNRS).

Doctoral school:
Mathematics and Computer Science, University of Bordeaux.

Research directions:

  • Designing ensembles of deep learning architectures to analyse 3D brain MRI volumes.
  • Mitigating the lack of labelled data through semi-supervised learning strategies.
  • Delivering decision-support tools for differential diagnosis of Alzheimer’s disease, frontotemporal dementia, and dementia with Lewy bodies.

Clinical partnerships:
Close collaboration with Bordeaux University Hospital and the Institute of Neurodegenerative Diseases to translate methodological advances into clinical practice.

Keywords (ADUM):

  • Magnetic resonance imaging
  • Deep learning
  • Transformers
  • Differential diagnosis
  • Neurodegenerative dementias
  • Semi-supervised learning

Resources: