The Matched Gabor Transform - A tool for adaptive phase extraction
-
1
Jena University Hospital - Friedrich-Schiller-University Jena, Institute for medical Statistics, Computer Sciences and Documentation, Germany
A core aspect of time-frequency analysis is always the question of time-frequency resolution. The latter is determined by the analysis method and has a crucial influence on the appearance of the result. Most methods, like Gabor transform or continuous wavelet transform, have parameters which have to be adapted to the data in order to obtain meaningful results. As the methods enforce a certain time-frequency resolution, the danger of interpreting effects of the methods instead of interpreting effects in the data is omnipresent [1].
The Matching Pursuit algorithm introduced by Mallet and Zhang in 1993 [2] decomposes a signal into dictionary atoms and builds up a pseudo Wigner-Ville distribution to provide a data-adaptive time-frequency resolution. The pseudo Wigner-Ville distribution is a pure power distribution and provides no phase information.
This drawback is adressed by the Matched Gabor Transform (MGT), recently published in [3]. It combines the decomposition into atoms of the Matching Pursuit algorithm with thereon adapted Gabor transforms. Accumulating the time-frequency planes yields the final result. By that, an data-adaptive phase extraction becomes possible.
We demonstrate the usage of the MGT on real-life EEG and MEG data of a photic driving experiment. Periodic optical stimuli lead to an entrainment of the alpha oscillation which can be seen by an amplitude increase as well as a phase locking after the stimulus onset.
Considering possible effects in the gamma frequency range, the method has a big advantage. As this area in the time-frequency plane can be contaminated by broad-band short-time saccadic artifacts [4], it is important to retain the natural characteristics of any observed activity to be able to distinguish oscillatory neuronal activity from saccadic interferences. Therefore, the MGT might be the future key tool for phase analysis in EEG and MEG analysis.
References
[1] T. Sauer, “Time-frequency analysis, wavelets and why things (can) go wrong,” Kognitive Neurophysiologie des Menschen - Human cognitive neurophysiology, vol. 4, no. 1, pp. 38–62, 2011.
[2] S. G. Mallat and Z. Zhang, “Matching pursuits with time-frequency dicitionaries,” IEEE Transactions on Signal Processing, vol. 41, no. 12, pp. 3397–3415, 1993.
[3] M. Wacker and H. Witte, „Adaptive Phase Extraction: Incorporating the Gabor Transform in the Matching Pursuit Algorithm,“ IEEE Transactions on bio-medical engineering, preprint online, 2011.
[4] S. Yuval-Greenberg and L. Y. Deouell, "The broadband-transient induced gamma-band response in scalp EEG reflects the execution of saccades," Brain Topogr, vol. 22, pp. 3-6, Jun 2009
Keywords:
data-adaptive,
phase extraction,
time-frequency analysis
Conference:
BC11 : Computational Neuroscience & Neurotechnology Bernstein Conference & Neurex Annual Meeting 2011, Freiburg, Germany, 4 Oct - 6 Oct, 2011.
Presentation Type:
Poster
Topic:
data analysis and machine learning (please use "data analysis and machine learning" as keyword)
Citation:
Wacker
M and
Witte
H
(2011). The Matched Gabor Transform - A tool for adaptive phase extraction.
Front. Comput. Neurosci.
Conference Abstract:
BC11 : Computational Neuroscience & Neurotechnology Bernstein Conference & Neurex Annual Meeting 2011.
doi: 10.3389/conf.fncom.2011.53.00223
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:
09 Aug 2011;
Published Online:
04 Oct 2011.
*
Correspondence:
Mr. Matthias Wacker, Jena University Hospital - Friedrich-Schiller-University Jena, Institute for medical Statistics, Computer Sciences and Documentation, Jena, Germany, matthias.wacker@mti.uni-jena.de