teanijarv/mribrew

python pipeline for dwi processing and analysis

mribrew

This repository contains scripts and tools for DWI data pre-processing and analysis. Its features include:

(1) converting raw DWI DICOM to NIfTI;

(2) pre-processing raw DWI data;

(3) running Mean Apparent Propagator MRI (MAPMRI) or Anatomically Constrained Tractography (ACT).

Literally all you need is three DICOM files (T1, DWI (dir-AP), and DWI (dir-PA)) and you can run all these scripts in order to get to the final results.

Getting Started

Overview

Main scripts

• mribrew_dcm2nifti.py: Script to convert DICOM to NIfTI.
• mribrew_dwi_processing.py: Script to run for pre-processing the raw DWI data.
• mribrew_dwi_mapmri.py: Script to run MAPMRI analysis using processed DWI data.
• mribrew_dwi_act.py: Script to run ACT using processed DWI data.
• mribrew_rsfmri_ebmconnectivity.py (adding more fMRI functionality in future): Script to run for functional connectivity analysis based on EBM stages with DK and Schaefer atlases using processed RSfMRI data.

Data Folder Structure

Follow the underlying folder structure for scripts to work without having to make modifications.

For mribrew_dcm2nifti.py, you need DICOM files in the data/dcm/. The output will be put to data/raw/ which will be used by mribrew_dwi_processing.py. The processed files will be put to data/proc/. Finally, mribrew_dwi_mapmri.py or mribrew_dwi_act.py will use these processed files and their output will be at data/res/.

data/
└── dcm/
    ├── sub-01/
    │   ├── Serie_03_t1_mprage_sag_p2_iso_1.0.zip
    │   ├── Serie_08_ep2d_diff_hardi_s2_pa.zip
    │   └── Serie_10_ep2d_diff_hardi_s2.zip
    ├── sub-02/
    │   └── ...
    └── ...
└── raw/
    ├── sub-01/
    │   └── anat/
    │       ├── T1w.json
    │       └── T1w.nii.gz
    │   └── dwi/
    │       ├── dir-AP_dwi.bval
    │       ├── dir-AP_dwi.bvec
    │       ├── dir-AP_dwi.json
    │       ├── dir-AP_dwi.nii.gz
    │       ├── dir-PA_dwi.bval
    │       ├── dir-PA_dwi.bvec
    │       ├── dir-PA_dwi.json
    │       └── dir-PA_dwi.nii.gz
    ├── sub-02/
    │   └── ...
    └── ...
└── proc/
    ├── sub-01/
    │   └── dwi/
    │       ├── eddy_corrected.nii.gz
    │       ├── brain_dwi_mask.nii.gz
    │       ├── gradChecked.bval
    │       └── gradChecked.bvec
    ├── sub-02/
    │   └── ...
    └── ...
└── res/
    ├── sub-01/
    │   └── mapmri/
    │       ├── sub-01/
    │       │   ├── sub-01_MSE.nii.gz
    │       │   ├── sub-01_QIV.nii.gz
    │       │   └── ...
    │       ├── sub-02/
    │       │   └── ...
    │       └── ...
    │   └── act/
    │       ├── sc_sift_1000000.csv
    │       └── tracks_sift_1000000.tck
    ├── sub-02/
    │   └── ...
    └── ...

Notes:

• Small delta and large delta are pre-defined in the mribrew_mapmri.py script - make sure to change that according to your dataset.
• misc/folder contains various files like acqp.txt and dcm2nii_config.json which you may need to change based on your data acquisition.

Running the Analysis

Run the mribrew_dwi_*.py script from the root directory of the project to begin the analysis:

python mribrew_dwi_*.py

Dependencies

Make sure you have all the necessary dependencies installed and the data is organized as per the above structure.

• FSL
• nibabel
• Nipype
• Graphviz
• MRtrix3
• DIPY
• nilearn

Contribution

If you'd like to contribute to this project or have any questions, please open an issue or submit a pull request.