MELD project – meld_focal_epilepsy

meld_focal_epilepsy – convert MELD dataset into a BIDS structure and send to the MELD team.

Overview

This program performs three steps :

1. DICOMS TO NIFTI : Goes over the meld_template folder and convert DICOMs into NIFTI. Results save as nifti format (.nii).
2. NIFTI TO BIDS : Data are copied over a temporary folder and defaced. Data are converted into BIDS format using dcm2bids.
3. COMPRESSION : BIDS folder is compressed and send over to the MELD team.

More detailled guidelines can be found in our Protocol 2 - MRI data protocol on Protocols.io.

Getting Started

Prerequisites

This program requires the following softwares and libraries :

• FSL 6.0 (installation guidelines)
• Anaconda (installation guidelines)

Installation

• Open a terminal and paste the following sentence :

  git clone https://github.com/MELDProject/meld_focal_epilepsy

  Press enter. Your package will be installed.
• Create the conda environment:

cd <path_to_meld_focal_epilepsy_github_folder>/meld_focal_epilepsy/scripts 
conda env create -f mfe_env.yml 

Your meld_focal_epilepsy package is ready to be used and you can activate it with :

conda activate mfe_env

Note : you will have to activate your environment every time you open a new terminal and you want to use the meld_focal_epilepsy package

Usage

1) Prepare the MELD data

Prepare your data following the MELD Focal Epilepsy architecture detailled at Protocol 2 - MRI data protocol \

a. Download the meld_focal_epilepsy_data folder from : https://figshare.com/s/763e50f4eb51a4f76f58
b. Copy the meld_template folder and rename with the MELD participant ID
c. Fill the folders with DICOMS/NIFTI from the respective sequences available (e.g T1w sequence acquired at 3T should go into the MELD_participantID/3T/T1/ folder)

2) Convert DICOMs into NIFTI

To launch the script, run

cd <path_to_meld_focal_epilepsy_github_folder>/meld_focal_epilepsy/scripts
python meld_bidsify_data_step1.py -d <path_to_meld_focal_epilepsy_data_folder>

where :

• <path_to_meld_focal_epilepsy_github_folder> : the path to where your meld_focal_epilepsy folder has been saved
• <path_to_meld_focal_epilepsy_data_folder> : the path to where your data are stored

3) Create the lesion mask

Create the lesion mask from the NIFTI T1 volume if not already done. Save the lesion mask as NIFTI into the right folder.

See Protocol 2 - MRI data protocol for more detailled guidelines on how to create the lesion mask

5) Convert NIFTI into BIDS format for the participants you which to send over.

You will need to create a list of participant in a text file e.g. MELD_H1_3T_P_0001 MELD_H1_3T_P_0002 MELD_H1_15T_P_0001 MELD_H1_15T_C_0002 To launch the script, run

cd <path_to_meld_focal_epilepsy_github_folder>/meld_focal_epilepsy/scripts
python meld_bidsify_data_step2.py -d <path_to_meld_focal_epilepsy_data_folder> -ids <list_participants.txt> [optional argument]

where :

• <path_to_meld_focal_epilepsy_github_folder> : the path to where your meld_focal_epilepsy folder has been saved
• <path_to_meld_focal_epilepsy_data_folder> : the path to where your data are stored
• <list_participants.txt> : the path to the txt file containing the list of participant's ids you wish to share with the MELD team
• optional arguments:
  • -njobs <number> : number of cpu to parallelise defacing (num, -1: all, -2: all minus 1)
  • --nodeface : use this flag if you do not need to deface your NIFTI images.
  • --nodel : use this flag if you do not want to the temporary folder to be deleted
  • -v : use this flag to print more warning/error information

Notes :

• This program has been only test for Linux users.

Example :

Activate Anaconda environment

conda activate mfe_env 

Run meld_bidsify scripts

 cd /home/documents/meld_focal_epilepsy/scripts
 python meld_bidsify_data_step1.py -d /home/documents/meld_focal_epilepsy_data

Then create lesion masks
Then run the second bidsify step

 cd /home/documents/meld_focal_epilepsy/scripts
 python meld_bidsify_data_step2.py -d /home/documents/meld_focal_epilepsy_data/ -ids /home/documents/meld_focal_epilepsy/participants/list_participants_batch_01072022.csv

Authors

Konrad WAGSTYL (UCL GOSICH, London)
Sophie ADLER (UCL GOSICH, London)
Mathilde RIPART (UCL GOSICH, London)\