Nos tutelles


   

   

nos reseaux sociaux


               

Rechercher




Accueil > Coin des étudiants > Stages Prolongeable par une thèse. > Tous les stages.

M2

Multiscale analysis of the cellular networks in bone and dentin.

publié le

Keywords : bioimaging ; cellular network ; multiscale structure ; image processing ; graph theory.

Morphometric studies of cell networks currently constitute an important challenge in bioimaging. The complexity of the analysis stems from : 1) the high degree of interconnectivity of the network and 2) its multiscale nature. Those aspects, which have received considerable attention in neurology, are still poorly characterized in mineralized tissues such as bone or tooth dentin despite considerable implications in the biomedical field (e.g. osteoporosis, tooth hypersensitivity).

This master project follows a series of studies undertaken at the LIPHY which allowed acquiring a unique expertise in confocal fluorescence microscopy and non-linear optics for the 3D characterization of the cellular network in bone3 and dentin4. Based on the image processing pipeline developped at the LIPHY, a MATLAB code was developed in 2017 in collaboration with the LARIS to analyze the multiscale connectivity based on graph theory.

The goal of this internship is to optimize the tools developed to analyze the cellular network on images compatible with a histological analysis routinely performed in laboratories dedicated to biological / biomedical analysis.

The goals of this internship are to :

  • acquire representative images of the cellular porosity in bone and tooth sections using confocal fluorescence and two-photon microscopy.
  • optimize the current image segmentation segmentation protocol using graph metrics.
  • Assess the potential of the codes developed for multiscale graph analysis for multifunctional studies of the bone and teeth samples.

This study relies, in first place, on the development of image processing methods for which a good knowledge is expected from the candidates. The code will be developed in Matlab and Python to ensure a widespread access. Some experience in microscopy would be appreciated (but is not mandatory).
Due to the positioning of the topic at the Physics/Medicine interface, the candidate will have to interact with actors from a broad range of disciplines. An open mind and good knowledge of english is therefore expected.
Depending on the candidate motivations, a continuation in a PhD will be encouraged and supported.

This project will run in the framework of a close collaboration with the LARIS lab (INRA / University of Angers) in connection with the MSSMat lab of Centrale-Supélec (Gif-sur-Yvette) and the biomedical unit INSERM U1033 of Lyon.

Contact :
Aurélien Gourrier, aurelien.gourrier@univ-grenoble-alpes.fr, 0476514729