Event Abstract

Back to Event

IBMA: An SPM toolbox for NeuroImaging Image-Based Meta-Analysis

Camille Maumet1* and Thomas E. Nichols1, 2
  • 1 University of Warwick, Warwick Manufacturing Group, United Kingdom
  • 2 University of Warwick, Department of Statistics, United Kingdom

Introduction
While most neuroimaging meta-analyses are based on peak coordinate data, the best practice method is an image-based meta-analysis that combines the effect estimates and the standard errors from each study [7]. Various efforts are underway to facilitate sharing of neuroimaging data to make such intensity-based meta-analysis possible (see, e.g. [4]).
When image data is available for each study, a number of approaches (see [6] for a review) have been proposed to perform such meta-analysis including combination of standardised statistics, just effect estimates or both effects estimates and their sampling variance. While the latter is the preferred approach in the statistical community [1], often only standardised estimates are shared, reducing the possible meta-analytic approaches.
In view of the increasing availability of image data for neuroimaging analyses, we introduce IBMA, a toolbox for SPM [8] providing a set of tools for image-based meta-analysis. The toolbox is freely available at: https://github.com/NeuroimagingMetaAnalysis/ibma.

Method
Using the IBMA toolbox, we studied six meta-analytic approaches based on:
- contrast estimates only: Random-effects General Linear Model (RFX GLM);
- contrast estimates and standard errors: Fixed-effects General Linear Model (FFX GLM);
- Z-statistic: Fisher’s [2], Stouffer [9], Mixed-effects (MFX) Stouffer [7];
- Z-statistic and sample size: Weighted-Z [5,10].
Out of these six approaches, two are random-effects methods (RFX GLM, Stouffer MFX) and therefore offers the possibility to deal with studies heterogeneity. The fixed-effects approaches are strictly only appropriate if the between-study variance is null.
Using 21 studies of pain in control subjects, we visually compared the results obtained at p< 0.05 FDR corrected using the six meta-analytic approaches. The reference results were computed with the best-practice analysis: a 3-level hierarchical model: level 1, subject FFX; level 2, study MFX; level 3: meta-analysis MFX, using FSL’s FLAME method [3].

Results and conclusion
Fig. 1 presents the detection obtained at p< 0.05 FDR corrected in a one-sample meta-analysis of pain using the IBMA toolbox. Further work will investigate the validity of each meta-analytic approach in the context of neuroimaging.

Figure 1
Image

Acknowledgements

This work was supported by the Wellcome Trust. We also gratefully acknowledge the use of this data from the Tracey pain group, FMRIB, Oxford.

References

[1] P. Cummings. 2004. Meta-analysis based on standardized effects is unreliable. Archives of pediatrics & adolescent medicine, 158(6):595–7.
[2] R.A. Fisher. 1932. Statistical Methods for Research Workers. Oliver and Boyd, Edinburgh.
[3] S.M. Smith, P.R. Bannister, C. Beckman, M. Brady, S. Clare, D. Flitney, P. Hansen, M. Jenkinson, D. Leibovici, B. Ripley, M. Woolrich, J. Zhang. NeuroImage. 2001. Fsl: new tools for functional and structural brain image analysis. 13 (6), 249.
[4] K. J. Gorgolewski, T. Yarkoni, S.S. Ghosh, R.A. Poldrack, J.-B. Poline, Y. Schwarz, D.S. Margulies. 2013. NeuroVault.org: A web database for sharing statistical parametric maps. Poster presented at 19th Annual Meeting of the Organization for Human Brain Mapping, Seattle, WA, USA.
[5] T. Liptak. 1958. On the combination of independent tests. Magyar Tud. Akad. Mat. Kutato Int. Kozl., 3:171–197
[6] J. Radua and D. Mataix-Cols. 2012. Meta-analytic methods for neuroimaging data explained. Biology of mood & anxiety disorders, 2(1):6.
[7] G. Salimi-khorshidi , S. M. Smith, J. R. Keltner, T. D. Wager and T. E. Nichols. 2009. Meta-analysis of neuroimaging data: a comparison of image-based and coordinate-based pooling of studies. NeuroImage, 45(3):810–23.
"Statistical Parametric Mapping." SPM -. Accessed April 03, 2014. http://www.fil.ion.ucl.ac.uk/spm/
[8] S. Stouffer, L. DeVinney, and E. Suchmen. 1949. The American Soldier: Adjustment During Army Life, volume 1. Princeton University Press, Princeton, NJ.
[9] D V Zaykin. 2011. Optimally weighted Z-test is a powerful method for combining prob- abilities in meta-analysis. Journal of evolutionary biology, 24(8):1836–41.

Keywords: Meta-analysis, Neuroimaging, MRI, fMRI methods, Statistics as Topic, Toolbox, SPM

Conference: Neuroinformatics 2014, Leiden, Netherlands, 25 Aug - 27 Aug, 2014.

Presentation Type: Poster, to be considered for oral presentation

Topic: Neuroimaging

Citation: Maumet C and Nichols TE (2014). IBMA: An SPM toolbox for NeuroImaging Image-Based Meta-Analysis. Front. Neuroinform. Conference Abstract: Neuroinformatics 2014. doi: 10.3389/conf.fninf.2014.18.00025

Copyright: The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers. They are made available through the Frontiers publishing platform as a service to conference organizers and presenters.

The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated.

Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed.

For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.

Received: 04 Apr 2014; Published Online: 04 Jun 2014.

* Correspondence: Dr. Camille Maumet, University of Warwick, Warwick Manufacturing Group, Coventry, United Kingdom, contact@camillemaumet.com