Vascular ageing is characterized by the occurrence of alterations in the elastic laminae present in the media of elastic arteries such as the aorta. Imaging of the mouse aorta by very high-resolution X-ray micro-tomography using synchrotron radiation makes it possible to observe these structures in their context, in 3D and at exceptional resolutions and scales of detail. The MODELAGE project aims to develop innovative approaches adapted to the analysis of this type of atypical data. Thus, it will aim to characterize as precisely as possible the natural ageing of these arteries and their environment in healthy mice, to model normal ageing and describe the discrete vascular alterations that characterize it. This objectified knowledge base, built on observations of samples from mice of varying ages from very young to very old, will then be used to detect and characterize alterations occurring at comparable ages in diabetic mice to discriminate between those due to ageing itself and those that may be associated with the pathology. These results will finally be used to predict vascular behavior both from the point of view of its possible ageing and its functionality. The MODELAGE project is a high-content biological image analysis project that will require the development of numerical methods and tools to meet the technological challenges associated with the mass and complexity of the data. Specific approaches to image analysis, machine learning and vascular simulation will take advantage of the latest high performance computing technologies.

The website for this ANR project, led by the MIME team, is now available for consultation : https://anr-modelage.gitlab.io/