|
Article information
2016 , Volume 21, ¹ 6, p.36-46
Klimova E.G., Medvedev S.B., Savostyanov A.N.
Algorithm for local filtering of low frequency and large amplitude noise in electroencephalograms data
Purpose. One of the problems in operation of electroencephalograms (EEG) data is the existence of “noise” in these data. The noise is connected both to a behavior of the patient, and to external influences. The purpose of this study is creation of an algorithm for filtering of low frequency and large amplitude noise arising from different sources in EEG data. At the same time one of requirements to the algorithm is elimination of the phase distortion in a wave. Methods. The algorithm is based on the Fourier analysis of data and can be applied locally in the set time interval. In the developed algorithm the approach to division of high-frequency and low-frequency fluctuations which is widely applied in the theory of turbulence is used. The algorithm is analog of the method of the sliding average. However it has a number of fundamental differences. The algorithm includes an averaging interval assessment procedure proceeding from the spectral analysis of data, and also criterion according to which the decision on removal of a low-frequency component of data (noise) in the set point is made. Results. Numerical experiments with real data of electroencephalograms are made. EEG registered on the basis of scientific research institute of physiology and fundamental medicine with the help of 128 channel encephalograph of Brain Products, Germany with bandpass range of 0.1-100 Hz, the frequency of digitization of a signal of 1000 Hz. Electrodes (126 EEG + VEOG) settled down according to the international diagram of 10-10 % with the reviewer Cz and a grounding electrode of AFz. Numerical experiments with real data of electroencephalograms (selection from 85 EEG) showed that the algorithm allows to adjust EEG data, cleaning low frequency noise and slightly changing wave phase. Conclusions. In this work the algorithm of a filtration of low-frequency noise in EEG data is proposed. Distinctive property of the offered algorithm is lack of distortion of a phase of a wave (within computational error of the method).
[full text]
Keywords: electroencephalogram, filtration of noise, spectral analysis
Author(s):
Klimova Ekaterina Georgievna
Dr. , Associate Professor
Position: Senior Research Scientist
Office: Federal Research Center for Information and Computational Technologies
Address: 630090, Russia, Novosibirsk, 6 Acad. Lavrentjev avenue
Phone Office: (383) 332 42 57
E-mail: klimova@ict.nsc.ru
SPIN-code: 4533-9357Medvedev Sergey Borisovich
Dr.
Position: Leading research officer
Office: Inctitute of Computational Technologies SB RAS
Address: 630090, Russia, Novosibirsk, Ac. Lavrentyev ave., 6
Phone Office: (383) 330-73-73
E-mail: serbormed@gmail.com
SPIN-code: 2140-1726Savostyanov Alexander Nikolaevich
Dr. , Associate Professor
Position: Leading research officer
Office: Scientific research institute of physiology and fundamental medicine, Novosibirsk state university
Address: 630117, Russia, Novosibirsk, Timakov St., 4
Phone Office: (383) 334-89-55
E-mail: Alexander.Savostyanov@gmail.com
References:
[1] Maryutina, T.M., Ermolaev, O.Yu. Vvedenie v psikhofiziologiyu [Introduction to psychophysiology]. Moscow: Flinta; 2004: 400. (In Russ.)
[2] Rangaraj M. Rangayyan Biomedical signal analysis: A case-study approach. Wiley; 2002: 552.
[3] Lee, T.-W., Girolami, M., Bell, A.J., Sejnowski, T.J. A unifying information-theoretic framework for independent component analysis. Computers and mathematics with applications. 2000; Vol. 39(11):1-21.
[4] Morbidi, F. , Garulli, A., Prattichizzo, D.,Rizzo, C. , Manganotti, P., Rossi, S. Off-line removal of TMS-induced artifacts on human electroencephalography by Kalman filter. Journal of Neuroscience Methods. 2007; 162(1-2):293-302.
[5] Knyazev, G.G. , Savostyanov, A.N., Bocharov, A.V., Tamozhnikov, S.S. , Saprigyn, A.E. Task-positive and task-negative networks and their relation to depression: EEG beamformer analysis. Behavioural brain research. 2016; 306(1):160-169. DOI: 10.1016/j.bbr.2016.03.033
[6] Biswa, l. B., Yetkin, F.Z., Haughton, V.M., Hyde, J.S. Functional connectivity in the motor cortex of resting human brain using echo-planar MRI. Magnetic resonance in medicine. 1995; 34(4):537-541.
[7] Savostyanov, A.N., Tsai, A.C., Liou, M., Levin, E.A., Lee, J.D., Yurganov, A.V., Knyazev, G.G. EEG correlates of trait anxiety in the stop-signal paradigm. Neuroscience Letters. 2009. 449(2):112-116.
[8] Yaglom, A.M. Correlation theory of stationary and related random functions. Supplementary Notes and References. New York: Springer; 1987: 258.
[9] Laykhtman, D.L. Dinamicheskaya meteorologiya [Dynamic meteorology]. Leningrad: Gidrometeoizdat; 1976: 606. (In Russ.)
[10] Kalitkin, N.N. Chislennye metody [Numerical methods]. Moscow: Nauka; 1978: 512. (In Russ.)
[11] Lloyd, E., Lederman, U., Ayvazyan, Tyurin, Yu.N. Spravochnik po prikladnoy statistike. V 3-kh tomakh [Handbook of applicable mathematics]. Moscow: Finansy I statistika; 1989: 510. ( In Russ.)
Bibliography link:
Klimova E.G., Medvedev S.B., Savostyanov A.N. Algorithm for local filtering of low frequency and large amplitude noise in electroencephalograms data // Computational technologies. 2016. V. 21. ¹ 6. P. 36-46
|
