ICA-Based EEG denoising: a comparative analysis of fifteen methods

• Autorzy: Albera L. , Kachenoura A. , Comon P. , Karfoul A. , Wendling F. , Senhadji L. , Merlet I.
• Języki publikacji: EN

Abstrakty

EN

Independent Component Analysis (ICA) plays an important role in biomedical engineering. Indeed, the complexity of processes involved in biomedicine and the lack of reference signals make this blind approach a powerful tool to extract sources of interest. However, in practice, only few ICA algorithms such as SOBI, (extended) InfoMax and FastICA are used nowadays to process biomedical signals. In this paper we raise the question whether other ICA methods could be better suited in terms of performance and computational complexity. We focus on ElectroEncephaloGraphy (EEG) data denoising, and more particularly on removal of muscle artifacts from interictal epileptiform activity. Assumptions required by ICA are discussed in such a context. Then fifteen ICA algorithms, namely JADE, CoM2, SOBI, SOBIrob, (extended) InfoMax, PICA, two different implementations of FastICA, ERICA, SIMBEC, FOBIUMJAD, TFBSS, ICAR3, FOOBI1 and 4- CANDHAPc are briefly described. Next they are studied in terms of performance and numerical complexity. Quantitative results are obtained on simulated epileptic data generated with a physiologically-plausible model. These results are also illustrated on real epileptic recordings.

Słowa kluczowe

EN

ICA; comparative analysis; EEG denoising

• Wydawca: Polska Akademia Nauk, Wydział IV Nauk Technicznych
• Czasopismo: Bulletin of the Polish Academy of Sciences. Technical Sciences
• Rocznik: 2012
• Tom: Vol. 60, nr 3
• Strony: 407--418
• Opis fizyczny: Bibliogr. 58 poz., rys., tab.

Twórcy

autor

Albera L.

autor

Kachenoura A.

autor

Comon P.

autor

Karfoul A.

autor

Wendling F.

autor

Senhadji L.

autor

Merlet I.

• INSERM, UMR 1099, F-35000 Rennes, France / Université de Rennes 1, LTSI, F-35000 Rennes, France

Bibliografia

• [1] B.W. Mcmenamin, A.J. Shackman, J.S. Maxwell, D.R. Bachhuber, A.M. Koppenhaver, L.L. Greischar, and R.J. Davidson, “Validation of ICA-based myogenic artifact correction for scalp and source-localized EEG”, NeuroImage 49 (3), 2416-2432 (2010).
• [2] T.-P. Jung, C. Humphries, T.W. Lee, S. Makeig, M.J. Mckeown, V. Iragui, and T.J. Sejnowski, “Extended ica removes artifacts from electroencephalographic recordings”, Advancesin Neural Information Processing Systems 10, 894-900 (1998).
• [3] T.P. Jung, S. Makeig, M. Westerfield, J. Townsend, E. Courchesne, and T.J. Sejnowski, “Independent component analysis of single-trial even-related potentials”, Second Int. Symposium onIndependent Component Analysis and Blind Signal Separation 1, 173-178 (1999).
• [4] S. Makeig, A.J. Bell, T.-P. Jung, and T.J. Sejnowski, “Independent component analysis of electroencephalographic data”, Advances in Neural Information Processing Systems 8, 145- 151 (1996).
• [5] R. Vigario and E. Oja, “BSS and ICA in neuroinformatics: from current practices to open challenges”, IEEE Reviews inBiomedical Engineering 1, 50-61 (2008).
• [6] R. Vigario, J. Sarela, V. Jousmaki, M. Hamalainen, and E. Oja, “Independent component approach to the analysis of EEG and MEG”, IEEE Trans. on Biomedical Engineering 47 (5), 589- 593 (2000).
• [7] L. Albera, P. Comon, L. Parra, A. Karfoul, A. Kachenoura, And L. Senhadji, “Biomedical applications”, in Handbook ofBlind Source Separation, eds. P. Comon and C. Jutten, Academic Press, London, 2010.
• [8] Y. Deville, C. Jutten, and R. Vigario, “Overview of source separation applications”, in Handbook of Blind Source Separation, eds. P. Comon and C. Jutten, Academic Press, London, 2010.
• [9] S. Olbrich, J. Jodicke, C. Sander, H. Himmerich, and U. Hegerl, “ICA-based muscle artefact correction of EEG data: What is muscle and what is brain?: Comment on McMenamin et al.”, NeuroImage 54 (1), 1-3 (2011).
• [10] S. Vorobyov and A. Cichocki, “Blind noise reduction for multisensory signals using ICA and subspace filtering, with application to eeg analysis”, Biological Cybernetics 86 (4), 293-303 (2002).
• [11] A. Kachenoura, L. Albera, and L. Senhadji, “S´eparation aveugle de sources en ing´enierie biom´edicale”, IRBM, Elsevier 28 (1), 20-34 (2007).
• [12] A. Kachenoura, L. Albera, L. Senhadji, and P. Comon, “ICA: a potential tool for BCI systems”, IEEE Signal Processing Magazine,Special Issue on Brain-Computer Interfaces 25 (1), 57-68 (2008).
• [13] F. Poree, A. Kachenoura, H. Gauvrit, C. Morvan, G. Carrault, and L. Senhadji, “Blind source separation for ambulatory sleep recording”, Trans. on Information Technology in Biomedicine 10 (2), 293-301 (2006).
• [14] A. Belouchrani, K. Abed-Meraim, J.F. Cardoso, and E. Moulines, “A blind source separation technique using second-order statistics”, IEEE Trans. on Signal Processing 45 (2), 434-444 (1997).
• [15] A. Ziehe and K.R. Muller, “TDSEP - an efficient algorithm for blind separation using time structure”, ICANN’98, Proc. 8thInt. Conf. on Artificial Neural Networks 1, 675-680 (1998).
• [16] T.W. Lee, M. Girolami, and T.J. Sejnowski, “Independent component analysis using an extended infomax algorithm for mixed sub-gaussian and super-gaussian sources”, Neural Computation 11 (2), 417-441 (1999).
• [17] S. Makeig, “EEGLAB: ICA toolbox for psychophysiological research”, WWW Site, Swartz Center for Computational Neuroscience,Institute of Neural Computation, University of SanDiego California, www.sccn.ucsd.edu/eeglab/ (2000).
• [18] P. Comon and C. Jutten, Handbook of Blind Source Separation, Academic Press, New York, 2010.
• [19] J. Karvanen and V. Koivunen, “Blind separation methods based on pearson system and its extensions”, Signal Processing, Elsevier 82, 663-673 (2002).
• [20] A. Belouchrani and A. Cichocji, “Robust whitening procedure in blind source separation context”, Electronics Letters 23 (24), 2050-2051 (2000).
• [21] C. Fevotte and C. Doncarli, “Two contributions to blind sources separation using time-frequency distributions”, IEEE Signalprocessing Letters 11 (3), 386-389 (2004).
• [22] J.F. Cardoso and A. Souloumiac, “Blind beamforming for nongaussian signals”, IEE Proceesings-F 140 (6), 362-370 (1993).
• [23] P. Comon “Independent analysis, a new concept?”, Signal Processing, Elsevier 36 (3), 287-314 (1994).
• [24] S. Cruces, L. Castedo, and A. Cichocki, “Novel blind source separation algorithms using cumulants”, Proc. on Int. Conf.on Acoustic, Speech and Signal Processing (ICASSP’2000) 1, 3152-3155 (2000).
• [25] S.A. Cruces-Alverez, A. Cichocki, and S. Amari, “On a new blind signal extraction algorithm: different criteria and stability analysis”, IEEE Signal Processing Letters 9 (8), 233-236 (2002).
• [26] A. Ferreol, L. Albera, and P. Chevalier, “Fourth order blind identification of underdetermined mixtures of sources (FOBIUM)”, ICASSP 03, 2003 IEEE Int. Conf. on Acoustics Speechand Signal Processing 1, 41-44 (2003).
• [27] A. Ferreol, L. Albera, and P. Chevalier, “Fourth order blind identification of underdetermined mixtures of sources (FOBIUM)”, IEEE Trans. on Signal Processing 53, 1254-1271 (2005).
• [28] L. Albera, A. Ferreol, P. Chevalier, and P. Comon, “ICAR: Independent Component Analysis using Redundancies”, ISCAS04, 2004 IEEE Int. Symposium on Circuits and Systems 1, CD-ROM (2004).
• [29] L. Albera, A. Ferreol, P. Chevalier, and P. Comon, “ICAR, a tool for blind source separation using fourth order statistics only”, IEEE Trans. on Signal Processing 53, 3633-3643 (2005).
• [30] L.D. Lathauwer, J. Castaing, and J.F. Cardoso, “Fourth-order cumulant-based blind identification of underdetermined mixtures”, IEEE Trans. on Signal Processing 55 (6), 2965-2973 (2007).
• [31] A. Karfoul, L. Albera, and L. De Lathauwer, “Canonical decomposition of even higher order cumulant arrays for blind underdetermined mixture identification”, SAM 08, Fifth IEEEWorkshop on Sensor Array and Multi-Channel Signal Processing 1, 501-505 (2008).
• [32] A. Karfoul, L. Albera, and L. De Lathauwer, “Iterative methods for the canonical decomposition of multi-way arrays: Application to blind underdetermined mixture identification”, SignalProcessing, Elsevier 91 (8), 1789-1802 (2011).
• [33] D. Cosandier-Rimele, J.M. Badier, P. Chauvel, and F. Wendling, “A physiologically plausible spatio-temporal model for depth-EEG signals recorded with intracerebral electrodes in human partial epilepsy”, IEEE Trans. on BiomedicalEngineering 3 (54), 380-388 (2007).
• [34] D. Cosandier-Rimele, I. Merlet, J.M. Badier, P. Chauvel, and F. Wendling, “The neuronal sources of EEG: modeling of simultaneous scalp and intracerebral recordings in epilepsy”, NeuroImage 42 (1), 135-146 (2008).
• [35] D. Cosandier-Rimele, I. Merlet, F. Bartolomei, J.M. Badier, and F. Wendling, “Computational modeling of epileptic activity: from cortical sources to eeg signals”, J. Clinical Neurophysiology 27 (6), 465-470 (2010).
• [36] J. Sarvas, “Basic mathematical and electromagnetic concepts of the biomagnetic inverse problems”, Physics in Medicine andBiology 32, 11-22 (1987).
• [37] T.Y. Young and T.W. Calvert, Classification, Estimation andPattern Recognition, American Elsevier Publishing Company, New York, 1974.
• [38] P. Mccullagh, Tensor Methods in Statistics, Chapman and Hall, London, 1987.
• [39] V.P. Leonov and A.M. Shiryaev, “On a method of calculation of semi-invariants”, Theory of Probability and its Applications 4, 319-29 (1959).
• [40] S. Amari, “Natural gradient works efficienttly in learning”, Neural Computation 10 (2), 251-276 (1998).
• [41] P. Comon, “Separation de melanges de signaux”, GRETSI 89,DouziŹme colloque sur le Traitement du Signal et des Images 1, 137-140 (1989).
• [42] E. Moreau, “Criteria for complex sources separation”, in EUSIPCO96, XIV Eur. Signal Proc. Conf. 1, 931-934 (1996).
• [43] J.F. Cardoso, “Blind signal separation: statistical principles”, Proc. IEEE 9 (10), 2009-2025 (1998).
• [44] A. Hyvarinen, “Fast and robust fixed-point algorithms for independent component analysis”, IEEE Trans. on Neural Networks 10 (3), 626-634 (1999).
• [45] A. Hyvarinen, “Jacobi angles for simultaneous diagonalization”, SIAM J. Matrix Analysis and Applications 17 (1), 161- 164 (1996).
• [46] X. Luciani and L. Albera, “Joint eigenvalue decomposition using polar matrix factorization”, in Latent Variable Analysisand Signal Separation LVA-ICA 2010, vol. 6365, pp. 612-619, Springer, Saint Malo, 2010.
• [47] P. Comon, X. Luciani, and A.L.F.D. Almeida, “Tensor decompositions, alternating least squares and other thales”, J.Chemometrics 23, CD-ROM (2009).
• [48] G.H. Golub and C.F.V. Loan, Matrix Computations, second edition, The Johns Hopkins University Press, Baltimore, 1989.
• [49] J.J. Rieta, F. Castells, C. Sanchez, V. Zarzoso, and J. Millet, “Atrial activity extraction for atrial fibrillation analysis using blind source separation”, IEEE Trans. on Biomedical Engineering 51 (7), 1176-1186 (2004).
• [50] V. Zarzoso and A. Hyvarinen, “Iterative algorithms”, in Handbookof Blind Source Separation, eds. P. Comon and C. Jutten, pp. 179-225, Academic Press, London, 2010..
• [51] G.H. Golub and C. Reinsch, “Singular value decomposition and least squares solutions”, Handbook for Automatic ComputationII, Linear Algebra, ed. J.H. Wilkinson, Springer, New York, 1970.
• [52] T.F. Chan, “An improved algorithm for computing the singular value decomposition”, ACM Transaction on MathematicalSoftware 8 (1), 72-83 (1982).
• [53] L.D. Lathauwer and J. Castaing, “Blind identification of underdetermined mixtures by simultaneous matrix diagonalization”, IEEE Trans. on Signal Processing 56, 3 (2008).
• [54] F. Wendling, J. Bellanger, F. Bartholomei, and P. Chauvel, “Relevance of nonlinear lumped-parameter models in the analysis of depth-eeg epileptic signals”, Biol. Cybern. 83, 367- 378 (2000).
• [55] S.I. Goncalves, J.C.D. Munck, J.P.A. Verbunt, F. Bijma, R.M. Heethaar, and F.L.D. Silva, “In vivo measurement of the brain and skull resistivities using an EIT-based method and realistic models for the head”, IEEE Trans. on BiomedicalEngineering 50 (6), 754-767 (2003).
• [56] G. Birot, L. Albera, F. Wendling, and I. Merlet, “Localisation of extended brain sources from eeg/meg: the exso-music approach”, Elsevier Neuroimage 56 (1), 102-113 (2011).
• [57] Z. Koldosky, P. Tichavsky, and E. Oja, “Efficient variant of algorithm FastICA for independent component analysis attaining the Cram´er-Rao lower bound”, IEEE Trans. Neural Netw. 17 (5), 1265-1277 (2006).
• [58] A. Yeredor, “Blind separation of Gaussian sources via secondorder statistics with asymptotically optimal weighting”, IEEESignal Processing Letters 7, 197-200 (2000).