INTEGRATED BAYESIAN ANALYSIS OF MEG AND FMRI

Summary

Principal Investigator: John George
Abstract: The goal of this proposal is to develop an integrated probabilistic approach to functional brain imaging using electromagnetic and hemodynamic techniques that capitalizes upon the strengths and minimizes the weaknesses of each technique alone. The fundamental rationale for attempting to integrate electromagnetic and hemodynamic imaging techniques is: (1) no single technique provides the spatial and temporal resolution needed for clinical and research applications; and (2) hemodynamic and electromagnetic techniques have complementary strengths and weaknesses that can be exploited in an integrated analysis. The mathematical basis for our probabalistic approach is Bayesian Inference. In contrast to most existing approaches to analysis of functional imaging data the results of our Bayesian inferential approach is not a single "best" estimate of brain activity according to some criterion, but rather estimates of the full probability distribution for parameters of interest. Furthermore, Bayesian inference can incorporate uncertain models, such as those of the hemodynamic response in fMRI or of the EEG forward model, which depends on uncertain conductivity profiles. We believe this Bayesian inference approach could significantly improve our ability to gain robust spatial-temporal information on neural activation from existing functional neural imaging modalities. In order to realize this we propose to, 1) develop a fully integrated analysis of fMRI and MEG data, 2) develop a probabilistic EEG foward model so that our MEG analysis can be used for EEG data, and 3) distribute, optimize, test and refine the spatial-temporal MEG analysis.
Funding Period: 1999-07-01 - 2008-06-30
more information: NIH RePORT

Top Publications

  1. ncbi Fast robust subject-independent magnetoencephalographic source localization using an artificial neural network
    Sung Chan Jun
    Biological and Quantum Physics Group, Los Alamos National Laboratory, Los Alamos, New Mexico 87545, USA
    Hum Brain Mapp 24:21-34. 2005
  2. ncbi Spatiotemporal Bayesian inference dipole analysis for MEG neuroimaging data
    Sung C Jun
    Biological and Quantum Physics Group, MS D454, Los Alamos National Laboratory, NM 87545, USA
    Neuroimage 28:84-98. 2005
  3. ncbi Improving source detection and separation in a spatiotemporal Bayesian inference dipole analysis
    Sung C Jun
    MS D454, Biological and Quantum Physics Group, Los Alamos National Laboratory, Los Alamos, NM 87545, USA
    Phys Med Biol 51:2395-414. 2006
  4. ncbi Spatiotemporal noise covariance estimation from limited empirical magnetoencephalographic data
    Sung C Jun
    MS D454, Applied Modern Physics Group, Los Alamos National Laboratory, Los Alamos, NM 87545, USA
    Phys Med Biol 51:5549-64. 2006
  5. ncbi Modeling spatiotemporal covariance for magnetoencephalography or electroencephalography source analysis
    Sergey M Plis
    Applied Modern Physics Group, Los Alamos National Laboratory, MS D454, Los Alamos, New Mexico 87545, USA
    Phys Rev E Stat Nonlin Soft Matter Phys 75:011928. 2007
  6. ncbi Probabilistic forward model for electroencephalography source analysis
    Sergey M Plis
    MS D454, Applied Modern Physics Group, Los Alamos National Laboratory, Los Alamos, NM 87545, USA
    Phys Med Biol 52:5309-27. 2007
  7. ncbi A generalized spatiotemporal covariance model for stationary background in analysis of MEG data
    S M Plis
    Dept of Comput Sci, New Mexico Univ, Albuquerque, NM 87545, USA
    Conf Proc IEEE Eng Med Biol Soc 1:3680-3. 2006
  8. pmc Bayesian brain source imaging based on combined MEG/EEG and fMRI using MCMC
    Sung C Jun
    Los Alamos National Laboratory, Los Alamos, NM 87545, USA
    Neuroimage 40:1581-94. 2008

Scientific Experts

  • Sung C Jun
  • Sergey M Plis
  • D M Ranken
  • D M Schmidt
  • S M Plis
  • Doug M Ranken
  • John S George
  • J S George
  • J ParĂ©-Blagoev
  • David M Schmidt
  • Petr L Volegov
  • C C Wood

Detail Information

Publications8

  1. ncbi Fast robust subject-independent magnetoencephalographic source localization using an artificial neural network
    Sung Chan Jun
    Biological and Quantum Physics Group, Los Alamos National Laboratory, Los Alamos, New Mexico 87545, USA
    Hum Brain Mapp 24:21-34. 2005
    ..We applied these methods to localize single dipole sources from MEG components isolated by blind source separation and compared the estimated locations to those generated by standard manually assisted commercial software...
  2. ncbi Spatiotemporal Bayesian inference dipole analysis for MEG neuroimaging data
    Sung C Jun
    Biological and Quantum Physics Group, MS D454, Los Alamos National Laboratory, NM 87545, USA
    Neuroimage 28:84-98. 2005
    ..Markov Chain Monte Carlo (MCMC) was used to sample the many possible likely solutions. The spatiotemporal Bayesian dipole analysis is demonstrated using simulated and empirical whole-head MEG data...
  3. ncbi Improving source detection and separation in a spatiotemporal Bayesian inference dipole analysis
    Sung C Jun
    MS D454, Biological and Quantum Physics Group, Los Alamos National Laboratory, Los Alamos, NM 87545, USA
    Phys Med Biol 51:2395-414. 2006
    ..The properties of this analysis in comparison to the previous one without active time range parameters are demonstrated through extensive studies using both simulated and empirical MEG data...
  4. ncbi Spatiotemporal noise covariance estimation from limited empirical magnetoencephalographic data
    Sung C Jun
    MS D454, Applied Modern Physics Group, Los Alamos National Laboratory, Los Alamos, NM 87545, USA
    Phys Med Biol 51:5549-64. 2006
    ..Finally, we present some localization results on median nerve stimulus empirical MEG data for our proposed noise covariance model...
  5. ncbi Modeling spatiotemporal covariance for magnetoencephalography or electroencephalography source analysis
    Sergey M Plis
    Applied Modern Physics Group, Los Alamos National Laboratory, MS D454, Los Alamos, New Mexico 87545, USA
    Phys Rev E Stat Nonlin Soft Matter Phys 75:011928. 2007
    ..Performance of ours and previous models is compared in source analysis of a large number of single dipole problems with simulated time courses and with background from authentic magnetoencephalography data...
  6. ncbi Probabilistic forward model for electroencephalography source analysis
    Sergey M Plis
    MS D454, Applied Modern Physics Group, Los Alamos National Laboratory, Los Alamos, NM 87545, USA
    Phys Med Biol 52:5309-27. 2007
    ..We conclude that the conductivity of the skull has to be either accurately measured by an independent technique, or that the uncertainties in the conductivity values should be reflected in uncertainty in the source location estimates...
  7. ncbi A generalized spatiotemporal covariance model for stationary background in analysis of MEG data
    S M Plis
    Dept of Comput Sci, New Mexico Univ, Albuquerque, NM 87545, USA
    Conf Proc IEEE Eng Med Biol Soc 1:3680-3. 2006
    ..Ways to estimate the value of the parameter controlling this tradeoff are also discussed...
  8. pmc Bayesian brain source imaging based on combined MEG/EEG and fMRI using MCMC
    Sung C Jun
    Los Alamos National Laboratory, Los Alamos, NM 87545, USA
    Neuroimage 40:1581-94. 2008
    ..The usefulness and feasibility of the method are verified through testing with both simulated and empirical data...