Multi-Scale Machine Learning Approach for Alzheimer Disease Progress Profiling

Théo Saulus

The infamous Alzheimer's disease is at the heart of numerous research, because of the number of patients suffering from it, and the lack of known cure at the moment. This is why researchers endeavor to better understand the process by which the disease evolves, and this at different scales. The project to which I will contribute within the Gentles Lab aims at exploiting multiscale data in order to better determine the progression of the disease, using machine learning and computational methods. A first aim of this work is to study genetic profiles of the brain regions affected, more or less severely, by the disease, in order to identify the specific genes differentiating these cells. A second objective is to characterize the spatial organization of the patients’ cells, which is known to be relevant for other degenerative diseases. These two aims together will allow a multi-scale profiling of progression of Alzheimer disease, taking advantage of modeling techniques previously developed in the lab for cancers. The expected outcome of this study is a better understanding of the underlying mechanisms of Alzheimer's disease, which will allow for further research to limit its impact or, hopefully, to treat it.


Academic Year
Area of Study