Description
Date depot: 18 novembre 2022
Titre: StroMap: Computational image analysis of the tumor stroma to understand its role in cancer metastasis
Directeur de thèse:
Jean-Christophe OLIVO-MARIN (Analyse d'images biologiques)
Encadrante :
Vannary MEAS-YEDID (Institut Pasteur (UMR3691))
Domaine scientifique: Sciences et technologies de l'information et de la communication
Thématique CNRS : Sciences de l’information et sciences du vivant
Resumé: Breast cancer is the most frequently occurring malignancy in women, and the 5-year survival rate and therapeutic options drops significantly for patients with metastases. The stromal microenvironment is increasingly recognized as a major factor promoting tumor growth, metastasis and resistance to therapy. The understanding of the stromal microenvironment and its modification require quantitative, reproducible and comprehensive analysis of the tumor stroma as an ecological system.
To address this question, we will leverage the dedicated imaging technics and the cutting-edge image analysis methods to map the stromal microenvironment to model the stromal organization.
This project aims at developing a powerful computational pipeline based on advanced Machine Learning for quantitative image analysis of cellular localization and composition within regions of interest and their relationships to help deciphering the role of stromal cell in breast cancer development and metastasis, with a focus on postpartum breast cancer.
We will develop
• New Deep Learning models to map and characterize tumor stromal entities (cells and structures).
• Efficient encoding of regions and structures of interest based on Geometrical algorithms for mining biological process modifications.
• Spatial statistics based on the framework of statistical pattern analysis to measure the interplay of tumor stromal actors (cell-cell and cell-region physical interactions).
Doctorant.e: Benimam Mounib Mohamed