NIPY
Neuroimaging tools for Python
Overview
Status
Members
Mailing list
Development
How to help
Overview
The NIPY project is currently in development. In due course it will be full-featured python program and development environment for analyzing functional imaging data.NIPY is from Neuroimaging In Python, and we pronounce it to rhyme with "SciPy" - in other words, both syllables rhyme with "pie". At least, that's how most of us pronounce it.
The previous name for this project was BrainPy. That proved to be unpopular with some of our team, so we changed it to NIPY. Go figure.
For the goals of the project, and other early project documents, please see the BrainPy site.
Status
The current code for NIPY is based on Jonathan Taylor's BrainSTAT, which was a port of Keith Worsley's fmristat program from Matlab to Python. You can download the most recent development code from here. NIPY has a full system for general linear modelling of FMRI data, and principal component analysis. Jonathan recently used NIPY/BrainSTAT and FSL to analyze the FIAC dataset.At the moment we are working on refactoring the codebase and adding more file formats. We are also starting to look into the spatial processing of fMRI data, exploring tools available from the SHFJ lab in Orsay and the ITK toolkit.
Further ahead is a full GUI, and integration with SPM, FSL, AFNI and VoxBo.
Members
Jonathan Taylor was the sole developer of the original BrainStat codebase and has contributed extensively to the effort to develop NIPY from the lessons learned his original project.Mark D'Esposito, Jarrod Millman, and the team at Berkeley are heavily involved with refactoring and extending the NIPY code base.
Matthew Brett is just getting going with contributing. He plans to port the MarsBaR ROI toolkit from matlab to python.
The team at Orsay with Jean-Baptiste Poline have developed a series of methods for fMRI. These are written in matlab or in C with python or matlab bindings (Alexis Roche, Philippe Ciuciu, Bertrand Thirion and many more). Yann Cointepas and colleagues wrote and maintain the BrainVisa environment and Denis Riviere has written the Anatomist display environment. These will be linked to the NIPY project.
Fernando Perez has contributed much to the design of the NIPY project, and is working with Francois Meyer on adding wavelet processing to NIPY. We love using his wonderful ipython interactive shell as a development tool.
John Hunter is the main author of Matplotlib, which is central to the display output in NIPY. Like the rest of the NIPY team, John is writing grants with us to support NIPY. Like Fernando, he is also advising us wisely about all things scientific python.
Federico Turkheimer and John Aston have written a grant to develop wavelet processing of neuroimaging data. This will start as a Python port of the Phiwave to Python, and will be a collaboration with Matthew, Fernando and Francois.
Tom Nichols plans to work with Jonathan Taylor on a toolkit for working on spatial multiple comparison correction which will integrate with NIPY.
How to help
We would be very interested to hear of anyone who would like to help us develop NIPY code.We are also getting to the stage where we need help in checking the documentation, code testing, and adding Wiki entries, feature suggestions, and discussions of code design. If you would like to help with this, please email Jarrod Millman or go ahead and signup for the NIPY developer's mailing list.