Skip to Main content Skip to Navigation
Theses

Imagerie adaptative en IRM : utilisation des informations de mouvements physiologiques pour l'optimisation des processus d'acquisition et de reconstruction

Abstract : Magnetic resonance imaging (MRI) is a relatively slow imaging technique. In the context of cardiac and abdominal imaging, patient motion is a major impediment that disturbs the spatial encoding process needed to form an image. Motion results in image deteriorations, called artifacts. These artifacts can take complex forms as this encoding occurs in an unusual space (Fourier/coil sensitivity hybrid space). Generally the patient is asked for a breathhold in order to minimize the influence of respiration, and the acquisition is synchronized to the electrocardiogram in order to handle cardiac contraction. These methods are imperfect and not always applicable, and therefore alternative approaches are desirable. We propose to integrate prior knowledge in the acquisition and reconstruction processes, based on a specially designed platform, developed to acquire and analyze physiological data during the MRI examination. Various solutions are investigated to implement this adaptive imaging, with special care to the correction of motion induced spatial encoding errors. For that purpose, we build a predictive model that allows elastic displacement fields in the field of view to be predicted, from linear combinations of signals provided by the platform. Then we define a generalized reconstruction framework in which predicted displacement data are included, leading to the reconstruction of a motion-compensated image. The hypotheses are analyzed, and the predictive model, as well as the proposed reconstruction methods, are validated on real cardiac and abdominal data from healthy volunteers, in 2D and 3D free breathing scans.
Document type :
Theses
Complete list of metadata

Cited literature [64 references]  Display  Hide  Download

https://hal.univ-lorraine.fr/tel-01748239
Contributor : Thèses Ul <>
Submitted on : Thursday, March 29, 2018 - 11:29:30 AM
Last modification on : Tuesday, October 13, 2020 - 10:46:39 AM
Long-term archiving on: : Friday, September 14, 2018 - 12:47:55 AM

File

SCD_T_2007_0107_ODILLE.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : tel-01748239, version 1

Citation

Freddy Odille. Imagerie adaptative en IRM : utilisation des informations de mouvements physiologiques pour l'optimisation des processus d'acquisition et de reconstruction. Autre. Université Henri Poincaré - Nancy 1, 2007. Français. ⟨NNT : 2007NAN10107⟩. ⟨tel-01748239⟩

Share

Metrics

Record views

80

Files downloads

1002