Development of an image processing pipeline for large-scale in vivo population studies of the brain microstructure

Details

Supervisors

Macq, Benoît

Faculty

Ecole polytechnique de Louvain

Degree label

Master [120] : ingénieur civil électricien, à finalité spécialisée

Abstract

Diffusion weighted magnetic resonance imaging (DW-MRI) is a rapidly evolving, non radiating and non invasive technique that allows to capture information on the brain microstructure through the restricted diffusion of water molecules. DW-MRI has seen a growing interest in the recent years motivating the acquisition of large multi-scanner multi-site data sets. The substantial acquisition time of this type of MRI sequence has also encouraged the extensive use of Echo Planar Imaging which suffers from additional artifacts and noise. Several tools have been developed in order to correct those individual problems but they come with the disadvantages of processing only one subject at a time and requiring different softwares making them cumbersome to use. This work presents and evaluates the performances of the ElikoPy library, a complete diffusion MRI processing pipeline that reduces common sources of artifact and captures information on the brain microstructure through multiple microstructural diffusion models. ElikoPy has been designed to deal with large databases and to be robust to different types of acquisitions. A particular attention has been paid to the correction of the motion artifacts and the handling of the challenges introduced by the new gradient cycling acquisition scheme.

Citations

• APA

Dessain, Q., & Simon, M. (2021). Development of an image processing pipeline for large-scale in vivo population studies of the brain microstructure. https://dial-mem.test.bib.ucl.ac.be/handle/2078.2/23126