Enhanced bootstrap method for statistical inference in the ICA model

Low complexity and stable bootstrap procedures are proposed for FastICA estimators.The developed procedures stem from the fast-and-robust bootstrap method, applicable for fixed-point estimating equations.We perform cost efficient and reliable bootstrap-based statistical inference on the mixing coefficients in the ICA model.Such statistical inferences are required to identify which sources contribute to a specific observed mixture signal.The developed techniques are utilized in identifying equipotential lines of the brain dipoles from EEG recordings. In this paper, we develop low complexity and stable bootstrap procedures for FastICA estimators. Our bootstrapping techniques allow for performing cost efficient and reliable bootstrap-based statistical inference in the ICA model. Performing statistical inference is needed to quantitatively assess the quality of the estimators and testing hypotheses on mixing coefficients in the ICA model. The developed bootstrap procedures stem from the fast and robust bootstrap (FRB) method [1], which is applicable for estimators that may be found as solutions to fixed-point (FP) equations. We first establish analytical results on the structure of the weighted covariance matrix involved in the FRB formulation. Then, we exploit our analytical results to compute the FRB replicas at drastically reduced cost. The developed enhanced FRB method (EFRB) for FastICA permits using bootstrap-based statistical inference in a variety of applications (e.g., EEG, fMRI) in which ICA is commonly applied. Such an approach has not been possible earlier due to incurred substantial computational efforts of the conventional bootstrap. Our simulation studies compare the complexity and numerical stability of the proposed methods with the conventional bootstrap method. We also provide an example of utilizing the developed bootstrapping techniques in identifying equipotential lines of the brain dipoles from electroencephalogram (EEG) recordings.

[1]  Abdelhak M. Zoubir,et al.  Bootstrap techniques for signal processing , 2004 .

[2]  William H. Press,et al.  Numerical Recipes 3rd Edition: The Art of Scientific Computing , 2007 .

[3]  Anthony C. Davison,et al.  Bootstrap Methods and Their Application , 1998 .

[4]  Ernst Fernando Lopes Da Silva Niedermeyer,et al.  Electroencephalography, basic principles, clinical applications, and related fields , 1982 .

[5]  Aapo Hyvärinen,et al.  Fast and robust fixed-point algorithms for independent component analysis , 1999, IEEE Trans. Neural Networks.

[6]  John S. Ebersole,et al.  EEG dipole modeling in complex partial epilepsy , 2005, Brain Topography.

[7]  Tzyy-Ping Jung,et al.  Independent Component Analysis of Electroencephalographic Data , 1995, NIPS.

[8]  Stephen M. Smith,et al.  Probabilistic independent component analysis for functional magnetic resonance imaging , 2004, IEEE Transactions on Medical Imaging.

[9]  Te-Won Lee.  Independent Component Analysis , 1998, Springer US.

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

[11]  Susan R. Wilson,et al.  Two guidelines for bootstrap hypothesis testing , 1991 .

[12]  Tianwen Wei,et al.  A Convergence and Asymptotic Analysis of the Generalized Symmetric FastICA Algorithm , 2014, IEEE Transactions on Signal Processing.

[13]  BOOTSTRAP HYPOTHESIS TESTING PROCEDURES. RESPONSE , 1993 .

[14]  Hyon-Jung Kim,et al.  On testing hypotheses of mixing vectors in the ICA model using fastica , 2011, 2011 IEEE International Symposium on Biomedical Imaging: From Nano to Macro.

[15]  Fernando Lopes da Silva,et al.  Comprar Niedermeyer's Electroencephalography, 6/e (Basic Principles, Clinical Applications, and Related Fields ) | Fernando Lopes Da Silva | 9780781789424 | Lippincott Williams & Wilkins , 2010 .

[16]  Aapo Hyvärinen,et al.  Testing Significance of Mixing and Demixing Coefficients in ICA , 2006, ICA.

[17]  Stefan Van Aelst,et al.  Fast and robust bootstrap , 2008, Stat. Methods Appl..

[18]  R. Zamar,et al.  Bootstrapping robust estimates of regression , 2002 .

[19]  Klaus Nordhausen,et al.  Deflation-Based FastICA With Adaptive Choices of Nonlinearities , 2014, IEEE Transactions on Signal Processing.

[20]  Esa Ollila,et al.  The Deflation-Based FastICA Estimator: Statistical Analysis Revisited , 2010, IEEE Transactions on Signal Processing.

[21]  Erkki Oja,et al.  Independent component approach to the analysis of EEG and MEG recordings , 2000, IEEE Transactions on Biomedical Engineering.

[22]  Aapo Hyvärinen,et al.  Validating the independent components of neuroimaging time series via clustering and visualization , 2004, NeuroImage.

[23]  R. Tibshirani,et al.  An introduction to the bootstrap , 1993 .

[24]  Saeid Sanei,et al.  EEG signal processing , 2000, Clinical Neurophysiology.

[25]  Erkki Oja,et al.  Performance analysis of the FastICA algorithm and Crame/spl acute/r-rao bounds for linear independent component analysis , 2006, IEEE Transactions on Signal Processing.

[26]  Visa Koivunen,et al.  Fast and robust bootstrap method for testing hypotheses in the ICA model , 2014, 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[27]  D. A. Dunnett.  Classical Electrodynamics , 2020, Nature.

[28]  Jens Haueisen,et al.  Dipole models for the EEG and MEG , 2002, IEEE Transactions on Biomedical Engineering.

[29]  L. Zhukov,et al.  Independent component analysis for EEG source localization , 2000, IEEE Engineering in Medicine and Biology Magazine.

[30]  Jarkko Ylipaavalniemi,et al.  Analyzing consistency of independent components: An fMRI illustration , 2008, NeuroImage.