PyNIT (Python NeuroImaging Toolkit)
Version: 0.1.5
PyNIT is a pipeline tool for in vivo neuroimaging, especially optimized for rodent, that runs on the Jupyter Notebook, an interactive research computing environment. This package provides hierarchically designed classes to help easily create and apply the pipelines using FSL, AFNI, and ANTs, which are the most used packaged in fMRI analysis. For consistent behavior, it is designed to work only with datasets that follow the BIDS (Brain Imaging Data Structure) guidelines.
Guideline of Project folder structure
- Project folder has to have three data class as subfolders : Data, Processing, Results
- Non-processed dataset need to be organized based on BIDS guideline as linked above, and have to be placed at subfolder 'Data'.
- After initiate project, 'Processing' and 'Results' folders will be automatically generated.
Example of project data structure
Single-session project
- Project_main_folder
- Data
- sub-01
- anat
- sub-01_T2w.nii.gz
- func<
- sub-01_task-optoP1_run-01_bold.nii.gz
- sub-01_task-optoP1_run-02_bold.nii.gz
- sub-01_task-optoP2_run-01_bold.nii.gz
- sub-01_task-optoP2_run-02_bold.nii.gz
- sub-01_task-rest_bold.nii.gz
- cbv
- sub-01_task-infusion_run-01_cbv.nii.gz
- anat
- sub-02
- anat ...
- sub-01
- Data
Multi-session project
- Project_main_folder
- Data
- sub-01
- ses-01
- anat
- sub-01_T2w.nii.gz
- func<
- sub-01_task-optoP1_run-01_bold.nii.gz
- sub-01_task-optoP1_run-02_bold.nii.gz
- sub-01_task-optoP2_run-01_bold.nii.gz
- sub-01_task-optoP2_run-02_bold.nii.gz
- sub-01_task-rest_bold.nii.gz
- cbv
- sub-01_task-infusion_run-01_cbv.nii.gz
- anat
- ses-02
- anat
- ...
- anat
- ses-01
- sub-02
- ses-01
- anat ...
- ses-02
- anat ...
- ses-01
- sub-01
- Data
License
PyNIT is licensed under the term of the GNU GENERAL PUBLIC LICENSE Version 3
Code
Install PyNIT with:
pip install pynit
And activate the widget feature with:
jupyter nbextension enable --py widgetsnbextension
Update with:
pip install --upgrade pynit
Author
The main author of PyNIT is currently SungHo Lee, please join us if you want to involve this project.