+1 Recommend
0 collections
      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      Machine learning for neuroimaging with scikit-learn


      Read this article at

          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.


          Statistical machine learning methods are increasingly used for neuroimaging data analysis. Their main virtue is their ability to model high-dimensional datasets, e.g., multivariate analysis of activation images or resting-state time series. Supervised learning is typically used in decoding or encoding settings to relate brain images to behavioral or clinical observations, while unsupervised learning can uncover hidden structures in sets of images (e.g., resting state functional MRI) or find sub-populations in large cohorts. By considering different functional neuroimaging applications, we illustrate how scikit-learn, a Python machine learning library, can be used to perform some key analysis steps. Scikit-learn contains a very large set of statistical learning algorithms, both supervised and unsupervised, and its application to neuroimaging data provides a versatile tool to study the brain.

          Related collections

          Most cited references25

          • Record: found
          • Abstract: found
          • Article: not found

          Distributed and overlapping representations of faces and objects in ventral temporal cortex.

          The functional architecture of the object vision pathway in the human brain was investigated using functional magnetic resonance imaging to measure patterns of response in ventral temporal cortex while subjects viewed faces, cats, five categories of man-made objects, and nonsense pictures. A distinct pattern of response was found for each stimulus category. The distinctiveness of the response to a given category was not due simply to the regions that responded maximally to that category, because the category being viewed also could be identified on the basis of the pattern of response when those regions were excluded from the analysis. Patterns of response that discriminated among all categories were found even within cortical regions that responded maximally to only one category. These results indicate that the representations of faces and objects in ventral temporal cortex are widely distributed and overlapping.
            • Record: found
            • Abstract: found
            • Article: not found

            Nipype: A Flexible, Lightweight and Extensible Neuroimaging Data Processing Framework in Python

            Current neuroimaging software offer users an incredible opportunity to analyze their data in different ways, with different underlying assumptions. Several sophisticated software packages (e.g., AFNI, BrainVoyager, FSL, FreeSurfer, Nipy, R, SPM) are used to process and analyze large and often diverse (highly multi-dimensional) data. However, this heterogeneous collection of specialized applications creates several issues that hinder replicable, efficient, and optimal use of neuroimaging analysis approaches: (1) No uniform access to neuroimaging analysis software and usage information; (2) No framework for comparative algorithm development and dissemination; (3) Personnel turnover in laboratories often limits methodological continuity and training new personnel takes time; (4) Neuroimaging software packages do not address computational efficiency; and (5) Methods sections in journal articles are inadequate for reproducing results. To address these issues, we present Nipype (Neuroimaging in Python: Pipelines and Interfaces; http://nipy.org/nipype), an open-source, community-developed, software package, and scriptable library. Nipype solves the issues by providing Interfaces to existing neuroimaging software with uniform usage semantics and by facilitating interaction between these packages using Workflows. Nipype provides an environment that encourages interactive exploration of algorithms, eases the design of Workflows within and between packages, allows rapid comparative development of algorithms and reduces the learning curve necessary to use different packages. Nipype supports both local and remote execution on multi-core machines and clusters, without additional scripting. Nipype is Berkeley Software Distribution licensed, allowing anyone unrestricted usage. An open, community-driven development philosophy allows the software to quickly adapt and address the varied needs of the evolving neuroimaging community, especially in the context of increasing demand for reproducible research.
              • Record: found
              • Abstract: found
              • Article: not found

              Information-based functional brain mapping.

              The development of high-resolution neuroimaging and multielectrode electrophysiological recording provides neuroscientists with huge amounts of multivariate data. The complexity of the data creates a need for statistical summary, but the local averaging standardly applied to this end may obscure the effects of greatest neuroscientific interest. In neuroimaging, for example, brain mapping analysis has focused on the discovery of activation, i.e., of extended brain regions whose average activity changes across experimental conditions. Here we propose to ask a more general question of the data: Where in the brain does the activity pattern contain information about the experimental condition? To address this question, we propose scanning the imaged volume with a "searchlight," whose contents are analyzed multivariately at each location in the brain.

                Author and article information

                Front Neuroinform
                Front Neuroinform
                Front. Neuroinform.
                Frontiers in Neuroinformatics
                Frontiers Media S.A.
                21 February 2014
                : 8
                : 14
                [1] 1Parietal Team, INRIA Saclay-Île-de-France Saclay, France
                [2] 2Neurospin, I2 BM, DSV, CEA Gif-Sur-Yvette, France
                [3] 3Institute of Computer Science VI, University of Bonn Bonn, Germany
                [4] 4Department of Computing, Imperial College London London, UK
                [5] 5Institut Mines-Telecom, Telecom ParisTech, CNRS LTCI Paris, France
                Author notes

                Edited by: Satrajit S. Ghosh, Massachusetts Institute of Technology, USA

                Reviewed by: Michael Hanke, Otto-von-Guericke-University, Germany; Eugene Duff, University of Oxford, UK

                *Correspondence: Alexandre Abraham, CEA Neurospin, Bât 145, Point Courrier 156, 91191 Gif-sur-Yvette, France e-mail: alexandre.abraham@ 123456inria.fr

                This article was submitted to the journal Frontiers in Neuroinformatics.

                Copyright © 2014 Abraham, Pedregosa, Eickenberg, Gervais, Mueller, Kossaifi, Gramfort, Thirion and Varoquaux.

                This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

                : 16 October 2013
                : 31 January 2014
                Page count
                Figures: 5, Tables: 1, Equations: 0, References: 37, Pages: 10, Words: 7004
                Methods Article

                machine learning,statistical learning,neuroimaging,scikit-learn,python
                machine learning, statistical learning, neuroimaging, scikit-learn, python


                Comment on this article