Ever encountered a file of which you weren’t sure what analysis steps it had gone through? Ever wanted to know what types of data you have available for a subject in one overview? Automatically document an analysis pipeline?

Provenance is meta-data that tracks the ‘history’ of a file, and niprov is a python program to create, store and publish provenance for brain imaging files.

To get started, install niprov using pip:

pip install niprov

Keep in mind that to open image files you may need other libraries, such as mne-python, nibabel or pydicom.

The next step is to look through your directories for image files:

provenance discover /my/data/dir

After which you can start looking at what you’ve collected:

provenance report

You can do the same in python code:

import niprov'.')
print( #this will return a list of dictionaries

Read on to see other features and options.

Get in touch if you have questions or suggestions, by submitting an issue on or via email to, or at tweet to @niprovenance


Indices and tables