Published online on June 14,2019
High resolution mapping of neural networks by optical microscopy in preclinical research
SUMMARY: The objective of this thesis work will be to develop original software tools to study whole brain neuron populations in preclinical research. The images will be acquired by optical microscopy at very high magnification to allow individualization at the scale of neurons. The counting and modelling of the cytoarchitecture of these cells will provide new information of great interest in healthy cases (anatomy, development) but also pathological cases (neurodegenerative diseases, neuronal death). Advanced image processing techniques will be used to conduct these analyses (machine learning, high performance computing). Finally, validation of the results obtained on multi-species data produced at CEA-MIRCen will be carried out by comparing the results obtained with the stereological reference technique (cell counting) and with anatomical segmentations (neuroanatomist expert, digital atlases).
MOTS-CLES : traitement de l'image, modélisation 3D, segmentation, machine learning
DIRECTEUR DE THESE : Thierry DELZESCAUX (thierry.delzescaux _at_ cea.fr)
ÉQUIPE D’ACCUEIL : CEA-MIRCen, LMN (Laboratoire des Maladies Neurodégénératives), Fontenay aux Roses - http://i2bm.cea.fr/drf/i2bm/Pages/mircen.aspx
FINANCEMENT : Programme Paris Region PhD (PhD2), Île-de-France
PARTENAIRE SOCIO-ECONOMIQUE : société NEOXIA (co-encadrant Cédric CLOUCHOUX)
DUREE DE LA THESE : 36 mois (date de début 10/2019)
UNIVERSITE / ECOLE DOCTORALE : Université Paris Saclay, ED EOBE (ED 575)