Desktop version

Home arrow Computer Science arrow Computational Diffusion MRI: MICCAI Workshop, Athens, Greece, October 2016

Source

Methods

Data and Pre-Processing

Imaging datasets for 17 unrelated subjects (10m, 7f aged 22-35) were randomly selected from the minimally pre-processed, 500-Subjects release of the Human Connectome Project (HCP). For a thorough description of the protocols and preprocessing pipelines refer to the HCP documentation [24-26]. In brief, data were collected on a Siemens 3T Skyra system. Each diffusion dataset comprised of 270 gradient directions, acquired evenly across three interleaved b-shells, b = 1000,2000 and 3000 s/mm2. An additional eighteen b=0 images were interleaved throughout the acquisition. The high angular and spatial (1.25 mm isotropic) resolution of the HCP datasets lends itself to investigations of grey matter diffusion where, partial volume effects and low anisotropy values are limiting factors.

The HCP pre-processing steps conducted prior to data release, include eddy current and motion correction, providing diffusion weighted images with good alignment and without major distortions. Therefore, further corrections to this end were not performed; however, HCP diffusion datasets suffer from subject specific gradient nonlinearities that were corrected for during the fitting procedure of the tissue model below.

 
Source
Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >

Related topics