IMAGeS team: IMages, leArning, Geometry and Statistics

TIBM: Data reconstruction

From IMAGeS team: IMages, leArning, Geometry and Statistics
Revision as of 15:25, 7 October 2016 by Noblet (talk | contribs)
Jump to navigation Jump to search


This research topic concerns the reconstruction of images from raw data acquired with various modalities. The contributions in this area were mainly realized through the project ERC FBrain for the reconstruction of fetal MRI, through the [ANR http://rhodes.unistra.fr Rhodes] for the three-dimensional tomographic reconstruction of macromolecules in cryo-electron microscopy (cryo-EM) and through works related to data inversion from polarimetric imaging.

  • The main problem encountered for tomographic reconstruction in cryo-EM is related to the fact that the projection angles are unknown and that images have a very low signal to noise ratio. To address this problem, a method for jointly estimating volume and angles of projections has been proposed [4-https://icube-publis.unistra.fr/4-FDNF13 FDNF13], [https: // icube-publis. unistra.fr/4-BBF15 4-BBF15], [4-https://icube-publis.unistra.fr/4-PBMT15 PBMT15], and methods for estimating projection parameters based on reduction techniques dimension 4-RBNM16 or moments 2-PBMT16 were developed.
  • Concernant la reconstruction d’IRM fœtales, il s’agit, à partir d’un ensemble de coupes acquises avec des séquences 2D ultra-rapide, de reconstruire un volume IRM 3D haute résolution, tout en corrigeant les mouvements du fœtus ainsi que les artefacts en intensité, que ce soit en imagerie morphologique 2-KHRS11 ou en IRM de diffusion 2-OKSD12, 2-FSKC14.
thumb

Thèses


→ Retour page TIBM