Parkinson's pipeline

Table of Contents

• Prerequisities
• Setup
• Script usage
• Resulting data

Prerequisities

The Parkinson's Pipeline has the following software requirements:

1. A 64-bit Linux Operating System.

2. Advanced Normalisation Tools (https://github.com/ANTsX/ANTs) for N4 bias field correction.

3. ROBEX brain extraction tool (https://www.nitrc.org/projects/robex).

4. Medical Image Registration ToolKit (https://github.com/BioMedIA/MIRTK).

5. MATLAB (we have tested version R2018a, but later versions and some older versions should work as well).

---

Setup

Data

The T1-weighted and neuromelanin-sentitive Nifti data to be processed must be located in a specific folder named Original inside ${GITHUB_ROOT}, where ${GITHUB_ROOT} refers to the folder where this repository has been cloned into. It must also adhere to the following naming convention:

${GITHUB_ROOT}
   |
   ---- Original
           |
           ---- sub-${subject_id}-${timepoint}_T1w.nii.gz
           ---- sub-${subject_id}-${timepoint}_${nm_seq_type}.nii.gz

• ${subject_id} is a unique identifier for each subject that can have any format.
• ${timepoint} is the timepoint of the data (for longitudinal studies) and must be a single integer starting from "1" for baseline data then "2" for first follow-up, and so on. If no longitudinal data is used, this can be any integer from "1" onwards.
• ${nm_seq_type} is the name of the neuromelanin-sensitive sequence this data represent and can have any format but must be setup as a sequence type (see Neuromelanin-sensitive sequence names).

Environmental variables

The script SetUpVariables.sh serves as a way for the user to setup the paths of the main software pre-requisites the pipeline has. Therefore, for the pipeline to work, the user has to edit this file to specify the specific software paths on their environment.

Neuromelanin-sensitive sequence names

The script SetUpNMTypes.sh serves as a way for the user to setup the names of the neuromelanin-sensitive sequences that will be processed. The user has to edit this file to specify the specific sequence names on their data. For every subject, only the files with ${nm_seq_type} names specified in the array will be processed. If a subject does not have data for any of the specified sequences, the pipeline will safely ignore it and proceed with the next sequence type.

---

Script usage

To run the pipeline, execute the run_pipeline.sh script. It will proceed with all subjects in the ${GITHUB_ROOT}/Original folder and create a ${subject_id}-${timepoint}.processed file for each of them upon completion. If a subject has already an associated .processed file, the pipeline will skip this subject and proceed with the next one.

---

Resulting data

The pipeline will store the resulting data using the following folder structure (which gets automatically created if it does not exist):

${GITHUB_ROOT}
   |
   ---- Registered
           |
           ---- dofs
           ---- MNI_space
           ---- MPRAGE_space

• dofs contains all the intermediate MIRTK transformations files.
• MPRAGE_space contains all the intermediate data in T1w structural space.
• MNI_space contains all the processed data in MNI152 space. This is the data that should be used for further analysis.

Inhomogeneity-corrected data

The processed neuromelanin-sensitive data in the MPRAGE_space and MNI_space folders will include a version with -corrected appended in the file name (before the .nii.gz extension). This corresponds to data that has been corrected for inhomogeneities within the brainstem area (MPRAGE_space folder) and then transformed into MNI152 space (MNI_space folder). We recommend to use the -corrected data within the MNI_space folder for further analyses.