PyMVPA: a Python Toolbox for Multivariate Pattern Analysis of fMRI Data
By Michael Hanke, Yaroslav O. Halchenko, Per B. Sederberg, Stephen Jose Hanson, James V. Haxby, Stefan Pollmann. 2009
486 citations recorded by the source corpus.
Decoding patterns of neural activity onto cognitive states is one of the central goals of functional brain imaging. Standard univariate fMRI analysis methods, which correlate cognitive and perceptual function with the blood oxygenation-level dependent (BOLD) signal, have proven successful in identifying anatomical regions based on signal increases during cognitive and perceptual tasks. Recently, researchers have begun to explore new multivariate techniques that have proven to be more flexible, more reliable, and more sensitive than standard univariate analysis. Drawing on the field of statistical learning theory, these new classifier-based analysis techniques possess explanatory power that could provide new insights into the functional properties of the brain. However, unlike the wealth of software packages for univariate analyses, there are few packages that facilitate multivariate pattern classification analyses of fMRI data. Here we introduce a Python-based, cross-platform, and open-source software toolbox, called PyMVPA, for the application of classifier-based analysis techniques to fMRI datasets. PyMVPA makes use of Python’s ability to access libraries written in a large variety of programming languages and computing environments to interface with the wealth of existing machine learning packages. We present the framework in this paper and provide illustrative examples on its usage, features, and programmability.
Subfields: Artificial Intelligence, Cognitive Neuroscience, Molecular Biology, Medical Imaging, Neuroscience
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- John-Dylan Haynes
- James V. Haxby
- Carlton Chu
- Christophe L. M. Phillips
- Alexandre Gramfort
- Janaina Mourão Miranda
- Yannick Schwartz
- Arturo Cárdenas-Blanco
In Cognitive Neuroscience
- James V. Haxby
- Gaël Varoquaux
- John-Dylan Haynes
- Andrew C. Connolly
- Maria Ida Gobbini
- Martin N. Hebart
- Bertrand Thirion
- J. Swaroop Guntupalli
In Molecular Biology
- James V. Haxby
- Andrew C. Connolly
- Alexandre Gramfort
- Dharshan Kumaran
- Gaël Varoquaux
- John Ashburner
- A. Maass
- Bertrand Thirion
In Medical Imaging
In Neuroscience
- James V. Haxby
- John-Dylan Haynes
- Alexandre Gramfort
- Martin N. Hebart
- Dharshan Kumaran
- Michael Eickenberg
- J. Swaroop Guntupalli
- Gaël Varoquaux
Citations in the corpus
- Strong Evidence for Pattern Separation in Human Dentate Gyrus (2016, methodology): “Trial estimates (64 per session, 256 in total) were analyzed for multivariate effects using PyMVPA 2.2.0 (Hanke et al., 2009).”
- A common, high-dimensional model of the representational space in human ventral temporal cortex (2011, methodology): “All further analyses were performed using MATLAB (version 7.8, MathWorks) and PyMVPA (Hanke et al., 2009; http://www.pymvpa.org).”
- Decoding neural representational spaces using multivariate pattern analysis. (2014, methodology): “…location of A a representational space can be investigated using searchlight analyses (Kriegeskorte et al. 2006, Chen et al. 2011, Oosterhof et al. 2011), and the topographic organization of that representation representational space into individual subjects’ topographies (Haxby et al. 2011).”
- Multivariate pattern analysis of fMRI: The early beginnings (2012, methodology): “He decided to develop PyMVPA, a Python-based MVPA toolbox, in collaboration with Yaroslav Halchenko (www.pymvpa.org) (Hanke et al., 2009), who was working with Steve Hanson at Rutgers University.”
- NeuCube: A spiking neural network architecture for mapping, learning and understanding of spatio-temporal brain data (2014, unknown): “…no ‘trace’ left after this type of learning has taken place, to help understand complex brain processes from data. fMRI data is a dynamic data that consists of a large number of voxels, measured over time, to represent a subject’s brain activity (e.g. Hanke et al., 2009; Haxby et al., 2011).”
- Nipype: A Flexible, Lightweight and Extensible Neuroimaging Data Processing Framework in Python (2011, unknown): “When combined with libraries such as Nibabel (neuroimaging data input and output), Numpy/Scipy (array representation and processing) and scikits-learn or PyMVPA (machine learning and data mining) the Function Interface provides means for rapid prototyping of complex data processing methods.”
- Machine learning for neuroimaging with scikit-learn (2014, methodology): “This example has already been widely analyzed (Hanson et al., 2004; Detre et al., 2006; O’Toole et al., 2007; Hanson and Halchenko, 2008; Hanke et al., 2009) and has become a reference example in matter of decoding.”
- PRoNTo: Pattern Recognition for Neuroimaging Toolbox (2013, methodology): “PyMVPA and Sci-kit Learn are sophisticated and flexible software packages primarily written in Python (a free and cross-platform programming language 5 ).”
- Statistical inference and multiple testing correction in classification-based multi-voxel pattern analysis (MVPA): Random permutations and cluster size control (2011, unknown): “, 2011; Pereira and Botvinick, 2011) and MVPA toolboxes (Hanke et al., 2009).”
- Assessing and tuning brain decoders: Cross‐validation, caveats, and guidelines (2016, methodology): “, 2001) as provided from the PyMVPA (Hanke et al., 2009) website –http://dev.”
- The Decoding Toolbox (TDT): a versatile software package for multivariate analyses of functional imaging data (2015, background): “%) contained the strings “PRoNTo,” “PyMVPA,” “MVPA toolbox” or “3dsvm,” which refer to previous MVPA-related software packages (see Comparison to Existing Packages of Multivariate Data Analysis).”