Abstract
We compared the performance of 22 algorithms for independent component analysis with the aim to find suitable algorithms for applications in the field of surface electrical brain activity analysis. The quality of the separation is assessed with four performance measures: a correlation coefficient based index, a signal-to-interference ratio, a signal-to-distortion-ratio and the computational demand. Artificial data are used consisting of typical electroencephalogram and evoked potentials signal patterns, e.g. spikes, polyspikes, sharp waves and spindles. We evaluate different noise scenarios and the influence of pre-whitening. The comparisons reveal considerable differences between the algorithms, especially concerning the computational load. Algorithms based on the time structure of the data set seem to have advantages in separation quality especially for sine-shaped signals. Derivates of FastICA and Infomax also attain good results. Our results can serve as a reference for selecting a task-specific algorithm to analyze a large number of signal patterns occurring in the surface electrical brain activity.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Abdi H (2007) Encyclopedia of measurement and statistics. Sage Publications Inc., Thousand Oaks
De Lucia M, Fritschy J et al (2008) A novel method for automated classification of epileptiform activity in the human electroencephalogram-based on independent component analysis. Med Biol Eng Comput 46(3):263–272. doi:10.1007/s11517-007-0289-4
Delorme A, Makeig S (2004) EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis. J Neurosci Methods 134(1):9–21. doi:10.1016/j.jneumeth.2003.10.009
DiPietroPaolo D, Müller H et al (2006) Noise reduction in magnetocardiography by singular value decomposition and independent component analysis. Med Biol Eng Comput 44(6):489–499. doi:10.1007/s11517-006-0055-z
Ebe M, Homma I (1994) Leitfaden für die EEG-Praxis. G. Fischer Verlag, Stuttgart
Fisher AC, El-Deredy W et al (2007) Removal of eye movement artefacts from single channel recordings of retinal evoked potentials using synchronous dynamical embedding and independent component analysis. Med Biol Eng Comput 45(1):69–77. doi:10.1007/s11517-006-0123-4
Giannakopoulos X, Karhunen J et al (1999) An experimental comparison of neural algorithms for independent component analysis and blind separation. Int J Neural Syst 9(2):99–114. doi:10.1142/S0129065799000101
Glass K, Frishkoff G et al (2004) A framework for evaluating ICA methods of artifact removal from multichannel EEG. Lect Notes Comput Sci 3195/2004:1033–1040
Göhler W (1986) Höhere Mathematik. VEB Deutscher Verlag für Grundstoffindustrie, Leipzig
Hyvarinen A, Karhunen J et al (2001) Independent component analysis. Wiley, New York
James CJ, Hesse CW (2005) Independent component analysis for biomedical signals. Physiol Meas 26(1):R15–R39
Knuth KH (1998) Maximum entropy and Bayesian methods. Springer, Boise
Knuth KH, Shah AS et al (2006) Differentially variable component analysis: identifying multiple evoked components using trial-to-trial variability. J Neurophysiol 95(5):3257–3276. doi:10.1152/jn.00663.2005
Krishnaveni V, Jayaraman S et al (2005) Comparison of independent component analysis algorithms for removal of ocular artifacts from electroencephalogram. Meas Sci Rev 5(2):67–78
Krishnaveni V, Jayaraman S et al (2006) Application of mutual information based least dependent component analysis (MILCA) for removal of ocular artifacts from electroencephalogram. Meas Sci Rev 1(1):63–74
Li Y, Powers D et al (2000) Comparison of blind source separation algorithms. World Scientific and Engineering Society Press, pp 18–21
Milanesi M, Martini N et al (2008) Independent component analysis applied to the removal of motion artifacts from electrocardiographic signals. Med Biol Eng Comput 46(3):251–261. doi:10.1007/s11517-007-0293-8
Nicolaou N, Nasuto SJ (2003) Comparison of temporal and standard independent component analysis (ICA) algorithms for EEG analysis. In: Tenth international conference on neural information processing (ICANN/ICONIP’03), pp 157–160
Rodenbeck A, Binder R et al (2006) A review of sleep EEG patterns. Part I: A compilation of amended rules for their visual recognition according to Rechtschaffen and Kales. Somnol Sleep Res Sleep Med 10(4):159–175
Stogbauer H, Kraskov A et al (2004) Least-dependent-component analysis based on mutual information. Phys Rev E Stat Nonlin Soft Matter Phys 70(6 Pt 2):66–123
Truccolo W, Knuth KH et al (2003) Estimation of single-trial multicomponent ERPs: differentially variable component analysis (dVCA). Biol Cybern 89(6):426–438. doi:10.1007/s00422-003-0433-7
Vincent E, Rémi G et al (2006) Performance measurement in blind audio source separation. IEEE Trans Audio Speech Lang Process 14(4). doi:10.1109/TSA.2005.858005
Weber H (1992) Einführung in die Wahrscheinlichkeitsrechnung und Statistik für Ingenieure. Teubner, Stuttgart
Wiklund U, Karlsson M et al (2007) Adaptive spatio-temporal filtering of disturbed ECGs: a multi-channel approach to heartbeat detection in smart clothing. Med Biol Eng Comput 45(6):515–523. doi:10.1007/s11517-007-0183-0
Xu W, Erdogmus D et al (2004) Independent component analysis and blind signal separation. Springer, Heidelberg
Acknowledgments
This work was in part supported by the German Ministry of Science (03IP605) and the German Research Council (DFG Ha 2899/7-1).
Author information
Authors and Affiliations
Corresponding author
Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Klemm, M., Haueisen, J. & Ivanova, G. Independent component analysis: comparison of algorithms for the investigation of surface electrical brain activity. Med Biol Eng Comput 47, 413–423 (2009). https://doi.org/10.1007/s11517-009-0452-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11517-009-0452-1