EBC Diffusion MRI Preprocessing
Scripted processing of the post-mortem chimpanzee diffusion MRI (dMRI) dataset acquired within the framework of the Evolution of Brain Connectivity (EBC) project.
Input
- DMRI EBC Dataset in Bruker PV 6.0.1 File structure
- High-Resolution FLASH Dataset in PV File structure
Output
- Processed dMRI Datasets
- Reorientation to MNI-Like Space
- Noise Bias Removal and MP PCA Denoising
- TOPUP and Motion Correction
- DTI Fit
Requirements
- FSL incl eddy_cuda
- MRTRIX
- ANTS
- Python 3 incl following libraries:
- Bru2Nii for Bruker Data Conversion
How to Use
- All processing variables can be modified in the 'SET_VARIABLES.sh' file.
- The preprocessing will only run in sequential order of the supplied files, as multiple processing steps rely on results from previous processing steps.
Application of Processing
- Clone the GitHub repository into the preprocessing folder
- Enter the absolute path of the raw Bruker files in the 'BRUKER_RAW_DIR' variable
- Run scripts '01_make_folders.sh', '02_make_nii.sh', '03_make_scanlist.sh' to generate the required folder structure, Nifti files and 'Scanlist.txt'
- Run the script '04_check_reorientation.sh'. If the reoriented image resembles an MNI oriented brain image, proceed. Otherwise, modify the variables noted under "Reorientation to MNI Space"
- Run script '05_flash_load_data.sh' to unpack and name the FLASH data
- Run script '06_diff_run_noise_characterization.sh' to analyze the acquired noise-map
- Run script '07_diff_run_topup.sh' to calculate the TopUp distortion correction. The TopUp warp field will not be applied directly to the data but combined with the other distortion fields in the dMRI processing step.
- Run script '08_diff_load_prep.sh' and '09_diff_run_preprocess.sh' to load adn preprocess the dMRI data, using the previously calculated results. Make sure to have CUDA activated.
- Run script '10_diff_run_bedpost.sh' to run bedpostX. Make sure to have CUDA activated.