BrainPy HBM abstract, 2005

This is the abstract describing the BrainPy / NIPY project from the HBM2005 conference.

BrainPy: an open source environment for the analysis and visualization of human brain data

Jonathan Taylor (1), Keith Worsley (2), Matthew Brett (3), Yann Cointepas (4), John Hunter (5), Jarrod Millman (3), Jean-Baptiste Poline (4), Fernando Perez (6)

  1. Dept. of Statistics, Stanford University, U.S.A.

  2. Dept. of Mathematics and Statistics, !McGill University, Canada

  3. Department of Neuroscience, University of California, Berkeley, U.S.A

  4. Service Hospitalier Frédéric Joliot, France

  5. Complex Systems Laboratory, University of Chicago, U.S.A.

  6. Department of Applied Mathematics, University of Colorado at Boulder, U.S.A.

Objective

What follows are the goals of BrainPy, a multi-center project to provide an open source environment for the analysis and visualization of human brain data built on top of python. While the project is still in its initial stages, packages for file I/O, script support as well as single subject fMRI and random effects group comparisons model are currently available.

Methods

Scientific computing has evolved over the last two decades in two broad directions. One, there has been a movement to the use of high-level interface languages that glue existing high-performance libraries into an accessible, scripted, interactive environment, eg IDL, matlab. Two, there has been a shift to open algorithms and software because this development process leads to better code, and because it more consistent with the scientific method.

Results & Discussion

The proposed environment includes the following:

  • We intend to provide users with an open source environment which is interoperable with current packages such as SPM and AFNI, both at a file I/O level and, where possible, interactively (e.g. pymat – calling matlab/SPM from python).

  • Read/write/conversion support for all major imaging formats and packages (SPM/ANALYZE, FSL, AFNI, MINC, NIFTI, and VoxBo

  • Low-level access to data through an interactive shell, which is important for developing new analysis methods, as well as high-level access through GUIs for specialized tasks using standard python tools.

  • Visualization of results using pre-existing tools such as BrainVisa, as well as support for development of new tools using VTK.

  • Support for MATLAB style numeric packages (Numarray) and plotting (matplotlib).

  • Support for EEG analysis including EEG/MEG/fMRI fusion analysis.

  • Support for spatio-temporal wavelet analysis (PhiWave)

Conclusions

BrainPy is an open-source environment for the analysis and visualization of neuroimaging data built on top of python.