# dwi-processing-* – Processing of DWI datasets¶

These pipelines perform a set of processing steps on corrected DWI datasets, currently:

• Diffusion tensor imaging (DTI) with extraction of DTI-based measures, namely the fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD) and radial diffusivity (RD). Then, normalization of the DTI-derived scalar maps (FA, MD, AD, RD) onto an FA-atlas with labelled tracts, and generation of TSV files containing a summary of the regional statistics (mean DTI-based measures) to ease subsequent statistical analyses. This is done by the dwi-processing-dti pipeline.
• (For multi-shell data only) Neurite Orientation Dispersion and Density Imaging (NODDI) with extraction of NODDI-based measures namely the neurite density index (NDI), the orientation dispersion index (ODI) ant the free water fraction (FWF). This is done by the dwi-processing-noddi pipeline.

To that aim, it mainly relies on the MRtrix [Tournier et al., 2012] software for the DTI aspects, NODDI Matlab Toolbox [Zhang et al., 2012] for NODDI aspects and on ANTS for the normalization aspects [Avants et al., 2008].

## Prerequisites¶

You need preprocessed DWI data prior to running any of these pipelines.

## Dependencies¶

If you only installed the core of Clinica, this pipeline needs the installation of ANTS, FSL and MRtrix (respectively ANTS, FSL and NODDI Matlab Toolbox) on your computer if you want to run the dwi-processing-dti (respectively dwi-processing-noddi) pipeline. You can find how to install these software packages on the third-party page.

## Running the pipeline¶

The processing pipelines can be run with the following command lines:

clinica run dwi-processing-dti caps_directory

clinica run dwi-processing-noddi caps_directory


where:

• caps_directory is the input/output folder containing the results in a CAPS hierarchy.

If you want to run the pipeline on a subset of your CAPS dataset, you can use the -tsv flag to specify in a TSV file the participants belonging to your subset.

## Outputs¶

### DTI-based outputs¶

Results are stored in the following folder of the CAPS hierarchy: subjects/sub-<participant_label>/ses-<session_label>/dwi/dti_based_processing/.

The main output files are:

• native_space/:
• <source_file>_space-[b0|T1w]_model-DTI_diffmodel.nii.gz: The diffusion tensor imaging (DTI) data of the subject.
• <source_file>_space-[b0|T1w]_[FA|MD|AD|RD].nii.gz: The DTI-based measures, namely the fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD) and radial diffusivity (RD).
• normalized_space/
• <source_file>_space-<space>_[FA|MD|AD|RD].nii.gz: The DTI-based measures registered to the space of an FA-atlas.
• atlas_statistics/
• <source_file>_space-<space>_map-[FA|MD|AD|RD]_statistics.tsv: TSV files summarizing the regional statistics on the labelled atlas <space>.

### NODDI-based outputs¶

Results are stored in the following folder of the CAPS hierarchy: subjects/sub-<participant_label>/ses-<session_label>/dwi/noddi_based_processing/.

The main output files are:

• native_space/:
• <source_file>_space-[b0|T1w]_model-NODDI_diffmodel.nii.gz: The Neurite Orientation Dispersion and Density Imaging (NODDI) data of the subject.
• <source_file>_space-[b0|T1w]_[FWF|NDI|ODI].nii.gz: The NODDI-based measures, namely namely the neurite density index (NDI), the orientation dispersion index (ODI) ant the free water fraction (FWF). <!--
• normalized_space/
• <source_file>_space-<space>_[FWF|NDI|ODI].nii.gz: The NODDI-based measures registered to the space of an ???????????????-atlas.
• atlas_statistics/
• <source_file>_space-<space>_map-[FWF|NDI|ODI]_statistics.tsv: TSV files summarizing the regional statistics on the labelled atlas <space>. -->

Atlases available for the DTI-based processing pipeline:

• JHUDTI81 [Hua et al., 2008; Wakana et al., 2007]: This atlas contains 48 white matter tract labels that were created by manually segmenting a standard-space average of diffusion MRI tensor maps from 81 subjects.
• JHUTracts[0|25|50] [Mori et al., 2005]. This atlas contains 20 white matter tract labels that were identified probabilistically by averaging the results of deterministic tractography run on 28 subjects. Several thresholds of these probabilistic tracts are proposed (0%, 25%, 50%).

## Describing this pipeline in your paper¶

Example of paragraph for the dwi-processing-dti pipeline:

These results have been obtained using the dwi-processing pipeline of Clinica. A diffusion tensor imaging (DTI) model was fitted to each voxel to calculate the fractional anisotropy (FA), mean diffusivity (MD), radial diffusivity (RD) and axial diffusivity (AD) maps using MRtrix [Tournier et al., 2012]. The FA map of each subject was then registered to the FA map of the JHU atlas template with the ANTs SyN algorithm [Avants et al., 2008], and the estimated non-linear deformation was applied to the MD, AD and RD maps to have, for each individual, all the DTI-based maps in the space of the JHU atlas.

We then assessed the integrity of a set of anatomical white matter tracts defined in the:

• (Description for JHUDTI81 atlas) DTI-81 white-matter atlas [Hua et al., 2008; Wakana et al., 2007]. This atlas contains 48 white matter tract labels that were created by manually segmenting a standard-space average of diffusion MRI tensor maps from 81 subjects.

• (Description for JHUTracts[0|25|50] atlas) JHU white-matter tractography atlas [Mori et al., 2005]. This atlas contains 20 white matter tract labels that were identified probabilistically by averaging the results of deterministic tractography run on 28 subjects. Several thresholds of these probabilistic tracts are proposed (0%, 25%, 50%).

The warping of this atlas to each individual subject provides a parcellation of the subject’s white matter into anatomical tracts. The integrity of the tracts was assessed by analyzing the average FA, MD, AD and RD in each tract. To that purpose, the scalar maps of each subject were put into correspondence with the FA-map in the atlas space using the ANTs SyN algorithm [Avants et al., 2008]. Finally, for each subject, the mean scalar value in each tract was computed for each DTI-based measure.

Example of paragraph for the dwi-processing-noddi pipeline:

These results have been obtained using the dwi-processing-noddi` pipeline of Clinica. A Neurite Orientation Dispersion and Density Imaging (NODDI) model was fitted to each voxel to calculate the neurite density index (NDI), the orientation dispersion index (ODI) ant the free water fraction (FWF) using the NODDI Matlab toolbox (https://www.nitrc.org/projects/noddi_toolbox) [Zhang et al., 2012]

Tip