A resampling approach to estimate the stability of one-dimensional or multidimensional independent components

When applying unsupervised learning techniques in biomedical data analysis, a key question is whether the estimated parameters of the studied system are reliable. In other words, can we assess the quality of the result produced by our learning technique? We propose resampling methods to tackle this question and illustrate their usefulness for blind-source separation (BSS). We demonstrate that our proposed reliability estimation can be used to discover stable one-dimensional or multidimensional independent components, to choose the appropriate BSS-model, to enhance significantly the separation performance, and, most importantly, to flag components that carry physical meaning. Application to different biomedical testbed data sets (magnetoencephalography (MEG)/electrocardiography (ECG)-recordings) underline the usefulness of our approach.

[1]  Jianbo Shi,et al.  Learning Segmentation by Random Walks , 2000, NIPS.

[2]  Lucas C. Parra,et al.  On-line Convolutive Blind Source Separation of Non-Stationary Signals , 2000, J. VLSI Signal Process..

[3]  S. T. Buckland,et al.  An Introduction to the Bootstrap , 1994 .

[4]  Joachim M. Buhmann,et al.  A Resampling Approach to Cluster Validation , 2002, COMPSTAT.

[5]  Lucas C. Parra,et al.  Convolutive blind separation of non-stationary sources , 2000, IEEE Trans. Speech Audio Process..

[6]  C. Jacobi,et al.  C. G. J. Jacobi's Gesammelte Werke: Über ein leichtes Verfahren, die in der Theorie der Sacularstorungen vorkommenden Gleichungen numerisch aufzulosen , 1846 .

[7]  A. J. Bell,et al.  Blind Separation of Event-Related Brain Responses into Independent Components , 1996 .

[8]  Harri Valpola,et al.  Bayesian Ensemble Learning for Nonlinear Factor Analysis , 2000 .

[9]  Pierre Comon,et al.  Independent component analysis, A new concept? , 1994, Signal Process..

[10]  Andreas Ziehe,et al.  Artifact Reduction in Magnetoneurography Based on Time-Delayed Second Order Correlations , 1998 .

[11]  Christian Jutten,et al.  Blind separation of sources, part I: An adaptive algorithm based on neuromimetic architecture , 1991, Signal Process..

[12]  David G. Stork,et al.  Pattern Classification (2nd ed.) , 1999 .

[13]  Motoaki Kawanabe,et al.  Estimating the Reliability of ICA Projections , 2001, NIPS.

[14]  J. Shao,et al.  The jackknife and bootstrap , 1996 .

[15]  Andrzej Cichocki,et al.  A New Learning Algorithm for Blind Signal Separation , 1995, NIPS.

[16]  Bernhard Schölkopf,et al.  Nonlinear Component Analysis as a Kernel Eigenvalue Problem , 1998, Neural Computation.

[17]  Andrew B. Kahng,et al.  Spectral Partitioning with Multiple Eigenvectors , 1999, Discret. Appl. Math..

[18]  Gabriel Curio,et al.  Magnetometry of injury currents from human nerve and muscle specimens using Superconducting Quantum Interferences Devices , 1999, Neuroscience Letters.

[19]  Gene H. Golub,et al.  Matrix computations , 1983 .

[20]  K D Wolff,et al.  Non-invasive neuromagnetic monitoring of nerve and muscle injury currents. , 1993, Electroencephalography and clinical neurophysiology.

[21]  G Curio,et al.  [Measuring para-DC biomagnetic fields of the head using a horizontal modulated patient cot]. , 1998, Biomedizinische Technik. Biomedical engineering.

[22]  Andreas Ziehe,et al.  Combining time-delayed decorrelation and ICA: towards solving the cocktail party problem , 1998, Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '98 (Cat. No.98CH36181).

[23]  Richard M. Everson,et al.  Independent Component Analysis: Principles and Practice , 2001 .

[24]  G. Curio,et al.  The effect of artifact rejection by signal-space projection on source localization accuracy in MEG measurements , 1999, IEEE Transactions on Biomedical Engineering.

[25]  Motoaki Kawanabe,et al.  ASSESSING RELIABILITY OF ICA PROJECTIONS - A RESAMPLING APPROACH , 2001 .

[26]  Risto J. Ilmoniemi,et al.  The signal-space projection (SSP) method , 1994 .

[27]  T. Sejnowski,et al.  MOVING-WINDOW ICA DECOMPOSITION OF EEG DATA REVEALS EVENT-RELATED CHANGES IN OSCILLATORY BRAIN ACTIVITY , 2000 .

[28]  Andreas Ziehe,et al.  TDSEP — an efficient algorithm for blind separation using time structure , 1998 .

[29]  Bart De Schutter,et al.  DAISY : A database for identification of systems , 1997 .

[30]  Erkki Oja,et al.  Independent Component Analysis for Identification of Artifacts in Magnetoencephalographic Recordings , 1997, NIPS.

[31]  Andreas Ziehe,et al.  An approach to blind source separation based on temporal structure of speech signals , 2001, Neurocomputing.

[32]  Andreas Ziehe,et al.  Independent component analysis of noninvasively recorded cortical magnetic DC-fields in humans , 2000, IEEE Transactions on Biomedical Engineering.

[33]  K. Müller,et al.  A comparison of ICA-based artifact reduction methods for MEG , 2000 .

[34]  S Makeig,et al.  Blind separation of auditory event-related brain responses into independent components. , 1997, Proceedings of the National Academy of Sciences of the United States of America.

[35]  Terrence J. Sejnowski,et al.  An Information-Maximization Approach to Blind Separation and Blind Deconvolution , 1995, Neural Computation.

[36]  D. Chakrabarti,et al.  A fast fixed - point algorithm for independent component analysis , 1997 .

[37]  J Vieth,et al.  SEF-MEG localization of somatosensory cortex as a method for presurgical assessment of functional brain area. , 1996, Electroencephalography and clinical neurophysiology. Supplement.

[38]  Andreas Ziehe,et al.  Unmixing Hyperspectral Data , 1999, NIPS.

[39]  Yair Weiss,et al.  Segmentation using eigenvectors: a unifying view , 1999, Proceedings of the Seventh IEEE International Conference on Computer Vision.

[40]  Erkki Oja,et al.  Subspace methods of pattern recognition , 1983 .

[41]  C. Jacobi.  Über ein leichtes Verfahren die in der Theorie der Säcularstörungen vorkommenden Gleichungen numerisch aufzulösen*). , 2022 .

[42]  G. Deco,et al.  An Information-Theoretic Approach to Neural Computing , 1997, Perspectives in Neural Computing.

[43]  Gabriel Curio,et al.  Non-invasive long-term recordings of cortical ‘direct current’ (DC–) activity in humans using magnetoencephalography , 1999, Neuroscience Letters.

[44]  E. Oja,et al.  Independent Component Analysis , 2013 .

[45]  Hagai Attias,et al.  Independent Factor Analysis , 1999, Neural Computation.

[46]  Jean-François Cardoso,et al.  Multidimensional independent component analysis , 1998, Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '98 (Cat. No.98CH36181).

[47]  L. Lathauwer,et al.  Fetal electrocardiogram extraction by source subspace separation , 1995 .

[48]  Antoine Souloumiac,et al.  Jacobi Angles for Simultaneous Diagonalization , 1996, SIAM J. Matrix Anal. Appl..

[49]  J. Cardoso,et al.  Blind beamforming for non-gaussian signals , 1993 .