New methods and strategies for the quantification of time-variant coupling and synchronization of oscillatory EEG/MEG activity of different frequencies with special focus on gamma-band activity

 

Overview

Figure 1: Sequence of PDC connectivity networks in the source space using current density estimates based on a 58 channel EEG measurement. (from Witte et al., Meth.Inf.Med. 48/2009,18-28).
Figure 2: EEG/MEG forward model in the time-frequency domain using a tight Gabor frame.

The analysis of coupling and synchronization of oscillatory EEG/MEG activity in the sensor or source space is an important research tool for the neuroscience. The aim of this project, is to enhance the required time-variant multi-variate analysis methods. For this purpose, methods for both the analysis of coupling and synchronization and the time-variant reconstruction of neural activity will be developed, optimized and tested with high-dimensional EEG/MEG data. Furthermore, we investigate the influence of signal preprocessing (filtering, signal decomposition, artifact rejection) on the results of the coupling and synchronization analysis in order to optimize the preprocessing strategy. Another main focus of the project is the development of source reconstruction algorithms forming the basis for the time-variant synchronization analysis in the source space. This algorithms require less signal preprocessing and work with low signal-to-noise ratio. All methods will be evaluated based on high-dimensional EEG/MEG data and compared to alternative approaches. Herein, the directed coupling of alpha-band and gamma-band activity, of high-frequency oscillations, and of reconstructed source activity is of main interest. During the project, we will analyze existing data from a photic-driving study and record new data to examine the conscious and unconscious perception of somatosensory stimuli.

Project partners

Publications & patents

Publications (Contributions to Conferences):

  • Gramfort A, Strohmeier D, Haueisen J, Hämäläinen MS, Kowalski M: Time-Frequency Mixed-Norm Estimates: Sparse M/EEG imaging with non-stationary source activations. Neuroimage, 70:410–422, 2013
  • Haueisen,J., Fleissig,K., Strohmeier,D.,  Elsarnagawy,T., Huonker,R., Liehr, M., Witte, O.W.: Reconstruction of quasi-radial dipolar activity using three-component magnetic field measurements. Clinical Neurophysiology, 123:1581-1585, 2012
  • Gramfort, A.; Strohmeier, D.; Haueisen, J.; Hämäläinen, M.; Kowalski, M.: Functional Brain Imaging with M/EEG Using Structured Sparsity in Time-Frequency Dictionaries. In: Székely, Gábor; Hahn, Horst (Editors): Information Processing in Medical Imaging (IPMI 2011), Lecture Notes in Computer Science, Vol. 6801, pp. 600-611, 2011
  • Strohmeier, D.; Gramfort, A.; Haueisen, J.; Hämäläinen, M.; Kowalski, M.: MEG/EEG source reconstruction based on Gabor thresholding in the source space. 8th International Symposium on Noninvasive Functional Source Imaging of the Brain and Heart (NFSI & ICBEM), pp.103-108, 2011
  • Wacker, M, Galicki, M, Putsche, P, Milde, T, Schwab, K, Haueisen, J, Ligges, C, Witte, H: A Time-Variant Processing Approach for the Analysis of Alpha and Gamma MEG Oscillations During Flicker Stimulus Generated Entrainment. IEEE Transactions on Biomedical Engineering, 58(11): 3069-3077, 2011
  • Strohmeier, D.; Halbleib, A.; Gratkowski, M.; Haueisen, J.: The Epsilon-Skew-Normal-Dictionary for the Decomposition of Single- and Multichannel Biomedical Recordings using Matching Pursuit Algorithms; Mediterranean Conference on Medical and Biological Engineering and Computing (MEDICON 2010), Chalkidiki, Greece, 2010
  • Halbleib, A.; Strohmeier, D.; Gratkowski, M.; Haueisen, J.: Source Localization Algorithm based on Topographic Matching Pursuit; Proceedings BMT 2010 / Jahrestagung der DGBMT 2010, Rostock, Germany, 2010
  • Graichen,U., Witte,H., Haueisen,J.: Analysis of induced components in electroencephalograms using a multiple correlation method. BioMedical Engineering OnLine, 2009, 8:21

Sponsorship

News

07/01/2014 Daniel Strohmeier receives Best Student Paper Award

At the 4th International Workshop on Pattern Recognition in Neuroimaging, which was organized by the Max Planck Institute for Intelligent Systems in Tübingen, Daniel Strohmeier wins the Best Student Paper Award.

Source reconstruction of auditory evoked fields using irMxNE and dSPM

At the 4th International Workshop on Pattern Recognition in Neuroimaging, which was organized by the Max Planck Institute for Intelligent Systems in Tübingen, Daniel Strohmeier wins the Best Student Paper Award.

The presented article:

Daniel Strohmeier, Jens Haueisen, Alexandre Gramfort: Improved MEG / EEG source localization with reweighted mixed norms.

presents a new method, iterative reweighted mixed norm estimate (irMxNE), for the reconstruction of focal sources in the brain from MEG and EEG data.The figure shows the result of a source localization of an auditory evoked field using irMxNE (blue dots) and dSPM in the right hemisphere of the brain.The sources reconstructed using irMxNE are located in the primary auditory centers and show good agreement with the maxima of the dSPM distribution.