Meeting Banner
Abstract #3635

Phantoms for Diffusion Simulations: Multi-Objective Differential Evolution for Realistic Numerical (MODERN) Phantoms.

Jonathan Rafael-Patino1, Thomas Yu1, Mariam Andersson2,3, Hans Martin Kjer2,3, Vedrana Andersen Dahl3, Alexandra Pacureanu4, Anders Bjorholm Dahl3, Tim B. Dyrby2,3, and Jean-Philippe Thiran1,5

1Signal Processing Lab (LTS5), École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland, 2Danish Research Centre for Magnetic Resonance, University Hospital Hvidovre, Hvidovre, Denmark, 3Department of Applied Mathematics and Computer Science, Technical University of Denmark, Kongens Lyngby, Denmark, 4European Synchrotron Radiation Facility, Genoble, France, 5Radiology Department, Centre Hospitalier Universitaire Vaudois and University of Lausanne, Lausanne, Switzerland

The following work presents a novel, generally applicable framework (MODERN) which takes, as input, statistics from histology and automatically generates biologically plausible numerical phantoms whose morphological features are optimized to match the input statistics. As a proof of concept, MODERN is used to generate three-dimensional geometrical meshes representing a bundle configuration of axons which can be immediately integrated into Monte Carlo simulations for diffusion MRI acquisitions. The obtained features of the optimised phantoms are shown to match those from the input values. The statistics used were obtained from axonal segmentations from synchrotron imaging.

This abstract and the presentation materials are available to members only; a login is required.

Join Here