Bayesian spatiotemporal model for detecting of active areas in brain for analyzing of fMRI data

authors:

avatar Nasrin Borumandnia , avatar Hamid Alavi Majd ORCID , * , avatar Farid Zayeri , avatar Ahmad Reza Baghestani , avatar Fariborz Faeghi , avatar Seyyed Mohammad Tabatabaei

Koomesh: Vol.19, issue 4; 845-851
published online: December 01, 2017
article type: Research Article
received: November 01, 2016
accepted: June 01, 2017

how to cite: Borumandnia N, Alavi Majd H, Zayeri F, Baghestani A R, Faeghi F, et al. Bayesian spatiotemporal model for detecting of active areas in brain for analyzing of fMRI data. koomesh. 2017;19(4):e152932. 

Abstract

Introduction: In recent years, Functional Magnetic Resonance Imaging (fMRI) has been highly regarded for determining activated areas of brain in different conditions. Statistical methods have a crucial role in analyzing fMRI data. These data have complicated 3-dimentional spatial and temporal correlation structures. Also, there is a time lapse between the stimulus onset and response, which is known as Hemodynamic Response Function (HRF). It is very important to consider the complex correlation structures and the behavior of HRF in statistical modeling. In present paper, a Bayesian spatiotemporal model is introduced that is applied to analyze fMRI data for detecting the activated areas of brain. Materials and Methods: Images related to 2-back task, obtained from a part of the My Connectome Project, was used that is implemented in Stanford University in 2015. The 3D spatiotemporal proposed model was fitted on the data, so that HRF is estimated for each voxel based on its data, separately, and complex correlation structures are also considered. FSL software was used for preprocessing of images and Matlab software was used for statistical modeling. Results: The results of statistical models show that some parts of inferior parietal and also frontal areas were activated by the task. Conclusion: A Bayesian spatiotemporal model was introduced as a suitable method for identifying activated areas in fMRI time series. Because of considering both of complex correlation structures and estimated HRF, our proposed model can be a perfect approach for analyzing of these data

References

  • 1.

    Poldrack RA, Mumford JA, Nichols TE. Handbook of functional MRI data analysis: Cambridge University Press; 2011.

  • 2.

    Handwerker DA, Ollinger JM, D'Esposito M. Variation of BOLD hemodynamic responses across subjects and brain regions and their effects on statistical analyses. Neuroimage 2004; 21: 1639-1651.

  • 3.

    Lu Y, Grova C, Kobayashi E, Dubeau F, Gotman J. Using voxel-specific hemodynamic response function in EEG-fMRI data analysis: An estimation and detection model. Neuroimage 2007; 34: 195-203.

  • 4.

    Marrelec G, Benali H, Ciuciu P, PlgriniIssac M, Poline JB. Robust Bayesian estimation of the hemodynamic response function in eventrelated BOLD fMRI using basic physiological information. Human Brain Mapping 2003; 19: 1-17.

  • 5.

    Lee KJ, Jones GL, Caffo BS, Bassett SS. Spatial Bayesian variable selection models on functional magnetic resonance imaging time-series data. Bayesian Anal 2014; 9: 699-732.

  • 6.

    Quirs A, Diez RM, Gamerman D. Bayesian spatiotemporal model of fMRI data. NeuroImage 2010; 49: 442-456.

  • 7.

    Woolrich MW, Jenkinson M, Brady JM, Smith SM. Fully Bayesian spatio-temporal modeling of fMRI data. IEEE Trans Med Imaging 2004; 23: 213-231.

  • 8.

    Zhang L, Guindani M, Versace F, Vannucci M. A spatio-temporal nonparametric Bayesian variable selection model of fMRI data for clustering correlated time courses. NeuroImage 2014; 95: 162-175.

  • 9.

    Zhang L, Guindani M, Vannucci M. Bayesian models for functional magnetic resonance imaging data analysis. Wiley Interdiscip Rev Comput Stat 2015; 7: 21-41.

  • 10.

    Zellner A. On assessing prior distributions and Bayesian regression analysis with g-prior distributions. Bayesian inference and decision techniques: Essays in Honor of Bruno De Finetti 1986; 6: 233-243.

  • 11.

    Harbison JI, Atkins SM, Dougherty MR, editors. N-back training task performance: Analysis and model. Proceedings of the Cognitive Science Society; 2011.

  • 12.

    Poldrack RA, Laumann TO, Koyejo O, Gregory B, Hover A, Chen MY, et al. Long-term neural and physiological phenotyping of a single human. Nat Commun 2015; 6: 8885.

  • 13.

    Penny W, Kiebel S, Friston K. Variational Bayesian inference for fMRI time series. NeuroImage 2003; 19: 727-741.

  • 14.

    Flandin G, Penny WD. Bayesian fMRI data analysis with sparse spatial basis function priors. NeuroImage 2007; 34: 1108-1125.

  • 15.

    Gssl C, Auer DP, Fahrmeir L. Bayesian spatiotemporal inference in functional magnetic resonance imaging. Biometrics 2001; 57: 554-562.

  • 16.

    Harrison LM, Green GG. A Bayesian spatiotemporal model for very large data sets. NeuroImage 2010; 50: 1126-1141.

  • 17.

    Smith M, Fahrmeir L. Spatial Bayesian variable selection with application to functional magnetic resonance imaging. J Am Stat Assoc 2007; 102: 417-431.

  • 18.

    Owen AM, McMillan KM, Laird AR, Bullmore E. Nback working memory paradigm: A metaanalysis of normative functional neuroimaging studies. Human Brain Mapping 2005; 25: 46-59.

  • 19.

    Ragland JD, Turetsky BI, Gur RC, Gunning-Dixon F, Turner T, Schroeder L, Chan R, Gur RE. Working memory for complex figures: an fMRI comparison of letter and fractal n-back tasks. Neuropsychology 2002; 16: 370.