Public | Automated Build

Last pushed: 3 years ago
Short Description
Short description is empty for this repo.
Full Description


This container calculates DTI parameters, based on diffusion MRI data.


fdata : str

The name of a nifti file with preprocessed diffusion data.

fbval : str

The name of a text-file with b-values in FSL format.

fbvec : str

The name of a text file with the b-vectors in FSL format.

fmask : str, optional

The name of a nifti file containing boolean mask of locations to
analyze. Default: no masking

fit_method: str, optional

Chooses the algorithm to use in fitting: {"WLS" | "OLS" | "NNLS" |

"RESTORE"}. Default: "WLS", which uses weighted least-squares (see [1]_ for
details). Choosing "RESTORE" will use an automated outlier-rejection
algorithm [2]_

.. [1] Comparison of bootstrap approaches for estimation of uncertainties of
DTI parameters. S. Chung, Y. Lu, H.G. Roland (2006). Neuroimage 33: 531-541.

.. [2] L.-C. Chang, D.K. Jones, C. Pierpaoli (2005). RESTORE: Robust estimation
of tensors by outlier rejection. MRM 53: 1088-1095


The mounted input folder should contain a metadata.json file with the following


Where fdata and fit_method are both optional.


root_tensor.nii.gz : file
A nifti file containing the 6 lower diagonal in the Nifti1 asymm format
(see and

root_{fa,md,ad,rd}: files
Nifti files containing the FA, Mean, Axial and Radial diffusivity, respectively.


To run this container use:

docker run --rm -it -v /path/to/data:/input -v /path/to/output/:/output arokem/dipy-dti

Where the folder /path/to/data/ should contain the metadata.json file,


This uses the dipy.reconst.dti module:

Docker Pull Command
Source Repository