Search & Find
DiaporamaDiaporamaDiaporamaDiaporamaDiaporamaDiaporamaDiaporamaDiaporamaDiaporamaDiaporamaDiaporamaDiaporamaDiaporamaDiaporamaDiaporamaDiaporamaDiaporamaDiaporama
Home
ICube Laboratory   >   Events : Accreditation to supervise Research : Contributions to the deformation and comparison of medical images

Accreditation to supervise Research : Contributions to the deformation and comparison of medical images

June 24, 2015
13:00
Illkirch - Pôle API - A207

Accreditation to supervise Research  : Vincent NOBLET

Team : MIV

Title : Contributions to the deformation and comparison of medical images

Abstract : My research interest in medical image processing mainly focuses on two top-ics, namely image registration on one hand and change detection in longitudinal sequences of di usion MRI on the other hand.

I started working on image registration during my PhD thesis on "Non-rigid registration of 3D brain images with topology preservation constraint". Then, I kept on working on this field through four main axes: (i) symmetrisation of registration and extension to groupwise registration, (ii) registration and warping of binary images, (iii) interpolation and reorientation of diffusion MRI images and (iv) simulation of cerebral atrophy.

Concerning the change detection in diffusion MRI sequences, a general frame-work based on the generalized likelihood ratio has rst been proposed to detect changes on different levels of representation of diffusion imaging, namely the apparent diffusion coefficient images, the diffusion tensor fields, and scalar im-ages characterizing diffusion properties. Alternative statistical tests based on a Log-Euclidean metric have then been considered to take into account the positive definite nature of diffusion tensor. Finally, two methods have been devised to detect focal and global changes along the main white matter fiber tracts.

Alongside these research activities and in the context of close collaboration with physicians and researchers in neuroscience, I brought my expertise to the investigation of various problems in neuroscience requiring appropriate image processing pipelines. These works mainly concern group comparison and correlation between clinical or cognitive data according and morphologicalor functional image data.

The members of the jury are Isabelle Bloch (Telecom Paris Tech, Paris), Michel Dojat (University Joseph Fourier, Grenoble), Jean-François Meder (Ste Anne Hospital, Paris), Stéphane Kremer (Strasbourg University hospital) and Fabrice Heitz (University of Strasbourg).

The presentation will take place on wednesday 24th june 2015 at 1.00pm in room A207 of the pôle API building in Illkirch.

À la une

Offers are available in the Job opportunities section of the ICube website or by clicking on the...

RSS Feeds

Flux RSS