Deformable image registration is a key technology for image analysis. Traditionally, approaches for deformable image registration have focused on well-defined mathematical models that allow inferring spatial deformations between image pairs via optimization. However, most recently a number of approaches have been proposed that replace optimization by training appropriate regression models from data. This talk will discuss some of these recent developments.

This presentation is part of Minisymposium “MS28 - Diffeomorphic Image Registration: Numerics, Applications, and Theory (2 parts)”
organized by: Andreas Mang (Department of Mathematics, University of Houston) , George Biros (Institute for Computational Engineering and Sciences, University of Texas at Austin) .

Authors:

Marc Niethammer (University of North Carolina at Chapel Hill)

Keywords:

image registration, machine learning