> Academic opportunities > PHD positions

Neural graph network for cellular interaction analysis

Technological challenge: Health and environment technologies, medical devices (learn more)

Department: Département Microtechnologies pour la Biologie et la Santé (LETI)

Laboratory: Laboratoire Systèmes d'Imagerie pour le Vivant

Start Date: 01-09-2022

Location: Grenoble

CEA Code: SL-DRT-22-0656

Contact: cedric.allier@cea.fr

The understanding of a cell culture requires the analysis of the cells but also more specifically the study of their interactions. Cells are constantly interacting with each other. It is these interactions that we must examine to understand the cell population as a whole. To date, digital microscopy has focused on the analysis of cells as isolated objects and there is little understanding on cellular interactions. To overcome this state of the art, we propose to conduct new analyses using 'Graph Neural Networks' coupled to digital microscopy. These graphs could characterize the central cells that have a strong structuring (homeostasis) or destructuring role (disease). In particular, cells with interactions different from the normal cell population could be identified as deviant or malignant. To establish these new 'Cellular Graph Neural Networks', we will rely on the lens-free video-microscopy developed since ten years in our laboratory which is able to provide large cell datasets necessary to perform this thesis work. The objective is to obtain 'Cellular Graph Neural Networks' able to predict the fate of cells within a population. The aim is to use these networks for cell classification, as a tool for cell culture simulation, to intuit new laws in cell biology, etc. The candidate should have a background in applied mathematics with a specialization in deep learning.

See all positions Download the offer (.zip)

Email Bookmark and Share