BIOMOD

 Global demographic growth has been accompanied by a progressive increase in the number of elderly people. In developed countries, life expectancy now exceeds 80 years. Among the main causes of illness are neurodegenerative diseases (NDD), and although the incidence of Alzheimer's disease seems to be falling significantly worldwide, the number of patients continues to rise in line with the ageing of the world's population, making it a major public health problem of the twenty-first century. These incapacitating and incurable diseases are the consequence of progressive degeneration and neuronal death, with complex and varied clinical pictures, the vast majority of which, notably cognitive NCDs, are associated with dysfunctions previously known as 'dementia' and 'minor/major cognitive impairment'. NCDs currently affect around 50 million people worldwide, with Alzheimer's disease accounting for 60-70% of cases. In France, around 900,000 people are affected and 200,000 new cases are diagnosed each year; 20% of people over 80 and 40% of people over 90 are affected. In addition, it is predicted that as the population ages, around 1.8 million people will be affected by Alzheimer's disease in France by 2050. In this project, we aim (1) to develop and apply statistical and mathematical models and algorithms for unsupervised classification (machine learning, ML), adapted to multimodal analysis, and (2) to exploit the connectomic approach (based on graph theory) enabling us to visualise NDMC phenotypes in the form of multimodal graphs (spatial representation of multimodal parameters and their interactions), complementary to the ML classification approach, based on a retrospective cohort including 269 patients.

Permanent·es : Monica Baciu
Postdoctorante : Elise Roger