Learning digital models of Alzheimer’s Disease progressionMS73

We will present mixed-effects models that relate the long-term history of phenomena based on short-term longitudinal observations. Complex dynamics involve non-linear interactions between features with various temporal characteristic. Riemannian geometry tools allow to model such spatiotemporal trajectories, describing average dynamics whose variations correspond to individual progression patterns. A stochastic version of the EM algorithm was used to estimated digital models of the Alzheimer’s Disease progression, at both a population and individual level.

This presentation is part of Minisymposium “MS73 - Mathematical Methods for Spatiotemporal Imaging (2 parts)
organized by: Chong Chen (LSEC, ICMSEC, Academy of Mathematics and Systems Science, Chinese Academy of Sciences) , Barbara Gris (Laboratoire Jacques-Louis Lions) , Ozan Öktem (KTH - Royal Institute of Technology) .

Igor Koval (Brain and Spine Institute, INRIA)
alzheimer disease progression, bayesian methods, mixed-effects model, nonlinear optimization