JP-MRPhys / DTI-ANALYSIS

Pipeline for advanced diffusion tensor image analysis, pre-processing, track based spatial statistics (TBSS) and tract visualisation

Example image (low resolution)

Requirement

1. Packages: NIPYPE, FSL, DICOM2NIFTI (22 June 2018 or later build, this avoids DTI re-ordering issues in philips data), ANTS, SPM V8
2. Configuration: Anaconda, MATLAB
3. DTI scan, b-vectors and b-values files, and flipped B0 acquistions to apply eddy current corrections

Stage 1 (PREPROCESSING)

Preprocessing (top-up and eddy current), DTI-FIT to generate FA and MD data using DTI scan, BVEC, BVALS, DTI_Bo_flipped (eddy current correction)

Stage 2 (Track-Based Spatial Statistics)

Using FA and MD values to do track based spatial stastistics (TBSS)

Stage 3 (Track Isolation: Not part of this script)

Trackvis: To do visulisation and isolation of individual tracks

Usage

Change the directories, and subject_list and filename template as per your data then run the script in terminal (and see if after are fews days)

python "yourfilename.py"

if necessary employ dcm2niix via utility script dcm2niix_conversion.py to convert dicom to nifti images

WIP

1. Diffussion kurtosis imaging
2. GPU support for eddy and DTI-FIT