Logo TU Ilmenau


Spatio-temporal decomposition with matching pursuit

Project description

Nowadays the technological development allows the construction of bioelectromagnetic measurement systems with several hundreds of measurement channels. In the traditional analysis of such multidimensional data often only time traces of selected single channels are considered. Another typical schema involves mapping analysis at single instants in time.

The aim of the project is the combination of the time-frequency analysis with the space mapping analysis. New methods, which are based on the Matching Pursuit decomposition combined with spatial analysis, are developed. For example it is possible by means of the Spatial Matching Pursuit (SMP) method to decompose multichannel data into parameterised spatiotemporal components. The SMP-atoms are extended atoms of the Matching Pursuit method. Besides time-frequency properties derived from Gabor atoms they have additional spatial properties, which describe the spatial distribution of multichannel data (see Figure). There are six different spatial modes. They are created using 2D Bessel functions and they model typical spatial distributions of multichannel bioelectromagnetic data.

The developed methods are used for analysis of EEG, MEG and MCG multichannel data. They are particularly useful for artefact rejection,  extraction of signal components of interest and time-frequency analysis of multichannel data.

Image 1: Adaptive time-frequency-space decomposition of MEG data with SMP method.
A – Butterfly plot of MEG data (31 channels); B – time-frequency-mode of MEG data; C – Butterfly plot of an extracted 600Hz component.

Project partner

Project partners were not deposited.

Publications & patents

  • Halbleib,A., Gratkowski,M., Schwab,K., Ligges,C., Witte,H., Haueisen,J.: Topographic analysis of engagement and disengagement of neural oscillators in photic driving: a combined EEG/MEG study. Journal of Clinical Neurophysiology, 29(1):33-41, 2012
  • Schneider R., Lau S., Kuhlmann L., Vogrin S.,  Gratkowski M., Cook M., Haueisen J. (2011), Matching pursuit based removal of cardiac pulse-related artifacts in EEG/fMRI World Academy of Science, Engineering and Technology, ISSN: 2010-376X, accepted
  • Halbleib A., Gratkowski M., Schwab K., Ligges C., Witte H., Haueisen J. (2011), Topographic analysis of engagement and disengagement of neural oscillators in photic driving: a combined EEG/MEG study, Journal of Clinical Neurophysiology, accepted
  • Lelic D., Gratkowski M., Hennings K., Drewes AM., Multichannel Matching Pursuit Validation and Clustering - a Simulation and Empirical Study, (2010) Journal of Neuroscience Methods, 196(1):190-200
  • Klemm,M., Haueisen,J., Ivanova,G.: Independent Component Analysis: Comparison of Algorithms for the Investigation of Surface Electrical Brain Activity. Medical & Biological Engineering & Computing, 47:413–423, 2009
  • Lelic D., Gratkowski M., Valeriani M.,Arendt-Nielsen L., Drewes A. (2009), Inverse Modeling on Decomposed Electroencephalographic Data: A Way Forward?, Journal of Clinical Neurophysiology, 26(4): 227-235
  • Ramon C., Holmes M., Freeman W.J., Gratkowski M., Eriksen J., Haueisen J. (2009), Power Spectral Density Changes During Covert Object Naming Tasks Measured With 256 Channel High Density EEG Recordings, Epilepsy and Behavior, 14(1): 54-59
  • Drewes A.M., Gratkowski M., Sami S.A., Dimcevski G., Funch-Jensen P., Arendt-Nielsen L. (2008) Is the Pain in Chronic Pancreatitis of Neuropathic Origin? Support from EEG Studies during Experimental Pain, World Journal of Gastroenterology, 14(25): 4020-4027
  • Gratkowski M., Haueisen J., Arendt-Nielsen L., CN Chen A. and Zanow F. (2008) Decomposition of Biomedical Signals in Spatial and Time-Frequency Modes, Method Inform Med, 47: 26-37
  • Gratkowski M., Haueisen J., Arendt-Nielsen L. and Zanow F. (2007) Topographic Matching Pursuit of spatio-temporal bioelectromagnetic data, Przegląd Elektrotechniczny (Electrical Review), ISSN 0033-2097, 83(11): 138-141
    Gratkowski M., Haueisen J., Arendt-Nielsen L., CN Chen A. and Zanow F. (2006) Time-Frequency Filtering of MEG Signals with Matching Pursuit, J Physiology-Paris, 99: 47-57


  • Gratkowski M., Schmidt S., Gießler F., Eiselt M., Güllmar D., Witte O., Haueisen J., Time-Frequency-Space Components of Somatosenory Evoked Potentials in Rats, 5th European Conference of the International Federation for Medical and Biological Engineering, Budapest, 14-18 September, 2011 
  • Gratkowski M., Lelic D., Haueisen J.: Inverse Modelling of SpatioTemporally Decomposed EEG data, Tagungsband Biosignalverarbeitung, Inovationen bei der Erfassung und Analyse bioelektrischer und biomagnetischer Signale Potsdam, 16.-18.07.2008, Universität Potsdam, ISBN 978-3-9810021-7-1, s. 79-82
  • Gratkowski M., Haueisen J.,Arendt-Nielsen L., Zanow F, Topographic Matching Pursuit of Spatio-Temporal Bioelectromagnetic Data, Proceedings of the XIV International Symposium on Theoretical Electrical Engineering ISTET'07, 20-23.7, 2007, Szczecin, Poland 
  • Gratkowski M, Haueisen J, Drewes A, Sami S, Dimcevski G: Spatial-Time-Frequency Analysis of Increased Theta Activity of The EEG in Chronic Pancreatitis Patients, Proceedings of Gemeinsame Jahrestagung der Deutschen, Österreichischen und Schweizerischen Gesellschaften für Biomedizinische Technik, 6-9 September, 2006, Zurich, Switzerland 
  • Gratkowski M, Haueisen J, Liehr M, Leder U, Goernig M, Arendt-Nielsen L: Time-frequency-space filtering distinguishes partly overlapping T- and U-waves in multichannel DC magnetocardiograms, Proceedings of 11th Congress of the International Society of Holter and Noninvasive Electrocardiology and the 32nd Congress of the International Society of Electrocardiology, 2-4 June, 2005, Gdansk, Poland, 321-323 
  • Gratkowski M, Haueisen J, Liehr  M, Leder U, Goernig M, Arendt-Nielsen L: Separation of T- and U-Waves by Means of Spatial-Time-Frequency Filtering, In: Proceedings of BMT 2005, the 39th Annual Congress of the German Society for Biomedical Engineering (DGBMT) within VDE, Nuremberg, Germany; Biomedizinische Technik (Biomedical Engineering)
  • Gratkowski M: Spatial-Time-Frequency Analysis of Biomagnetic Data, Proceedings 26. Symposium des Arbeitskreises Mathematik in Forschung und Praxis Im Physikzentrum Bad Honnef 28.02-01.03.2005 
  • Gratkowski M., Haueisen J.: Analysis of EEG and MEG Signals with Matching Pursuit, 49. Jahrestagung der Deutschen Gesellschaft für Klinische Neurophysiologie und funktionelle Bildgebung, Klinische Neurophysiologie 35: 146 , DGKN September 15-19, Jena, 2004 
  • Gratkowski M.,Haueisen J., Schack B., Arendt-Nielsen L., Chen A.C.N., Zanow F.: Time-Frequency Analysis of Single Trial MEG Data using Matching Pursuit, Proceedings of the 4th International Symposium on Noninvasive Functional Source Imaging within the human brain and heart, Chieti – Italy, September 10-13, 2003 
  • Gratkowski M., Haueisen J., Schack B., Arendt-Nielsen L., Chen A., Zanow F.: Comparison of Wavelet Transform and Matching Pursuit in the Analysis of EEG and MEG Signals, In: Boenick, U., Bolz, A. (Hrsg.): Biomedizinische Technik: Beiträge zur gemeinsamen Jahrestagung der Deutschen, der Österreichischen und der Schweizerischen Gesellschaften für Biomedizinische Technik, September 24-27. Band 48, Ergänzungsband 1: 186-187. Salzburg, 2003
  • Gratkowski, M., Haueisen, J., Schack, B.: Time-Frequency Filtering with Matching Pursuit, 48. Jahrestagung der GMDS 2003, Muenster,  Germany, 14-18. September 2003, Informatik, Biometrie und Epidemiologie in Medizin und Biologie (2003), 584 – 585