Adaptive spectral tracking for coherence estimation: the z-tracker

OBJECTIVE A major challenge in non-stationary signal analysis is reliable estimation of correlation. Neurophysiological recordings can be many minutes in duration with data that exhibits correlation which changes over different time scales. Local smoothing can be used to estimate time-dependency, however, an effective framework needs to adjust levels of smoothing in response to changes in correlation. APPROACH Here we present a novel data-adaptive algorithm, the z-tracker, for estimating local correlation in segmented data. The algorithm constructs single segment coherence estimates using multi-taper windows. These are subject to adaptive Kalman filtering/smoothing in the z-domain to construct a local coherence estimate for each segment. The error residual for each segment determines the levels of process noise, allowing the filter to adapt rapidly to sudden changes in correlation while applying greater smoothing to data where the correlation is consistent across segments. The method is compared to wavelet coherence, calculated using orthogonal Morse wavelets. MAIN RESULTS The performance of the z-tracker is quantified against Morse wavelet coherence using a mean square deviation (MSD) metric. The z-tracker has significantly lower MSD than the wavelet estimate for time-varying coherence over long time scales (∼10-20 s), whereas the wavelet has lower MSD for coherence varying over short time scales (∼1-2 s). The z-tracker also has a lower MSD for slowly varying coherence with occasional step changes. The method is applied to detect changes in coherence in paired LFP recordings from rat prefrontal cortex and amygdala in response to a pharmacological challenge. SIGNIFICANCE The z-tracker provides an effective and efficient method to estimate time varying correlation in multivariate data, leading to better characterisation of neurophysiology signals where correlation is subject to slow modulation over time. A number of suggestions are included for future refinements.

[1]  G. Wittum,et al.  Adaptive filtering , 1997 .

[2]  N. R. Goodman,et al.  GAUSSIAN APPROXIMATIONS TO THE DISTRIBUTION OF SAMPLE COHERENCE , 1965 .

[3]  Andrew T. Walden,et al.  A Statistical Study of Temporally Smoothed Wavelet Coherence , 2010, IEEE Transactions on Signal Processing.

[4]  A. Izenman,et al.  Fourier Analysis of Time Series: An Introduction , 1977, IEEE Transactions on Systems, Man and Cybernetics.

[5]  M. B. Priestley,et al.  Non-linear and non-stationary time series analysis , 1990 .

[6]  B A Conway,et al.  Organization of common synaptic drive to motoneurones during fictive locomotion in the spinal cat , 2005, The Journal of physiology.

[7]  J. R. Rosenberg,et al.  The Fourier approach to the identification of functional coupling between neuronal spike trains. , 1989, Progress in biophysics and molecular biology.

[8]  Joerg F. Hipp,et al.  Time-Frequency Analysis , 2014, Encyclopedia of Computational Neuroscience.

[9]  M. B. Priestley,et al.  WAVELETS AND TIME‐DEPENDENT SPECTRAL ANALYSIS , 1996 .

[10]  C. Torrence,et al.  A Practical Guide to Wavelet Analysis. , 1998 .

[11]  Demba Ba,et al.  Robust spectrotemporal decomposition by iteratively reweighted least squares , 2014, Proceedings of the National Academy of Sciences.

[12]  A. Nuttall,et al.  Statistics of the estimate of the magnitute-coherence function , 1973 .

[13]  D. Halliday,et al.  Systemic administration of the benzodiazepine receptor partial inverse agonist FG‐7142 disrupts corticolimbic network interactions , 2007, Synapse.

[14]  Sofia C. Olhede,et al.  Higher-Order Properties of Analytic Wavelets , 2008, IEEE Transactions on Signal Processing.

[15]  M. Arnold,et al.  Instantaneous multivariate EEG coherence analysis by means of adaptive high-dimensional autoregressive models , 2001, Journal of Neuroscience Methods.

[16]  Guanrong Chen,et al.  Introduction to random signals and applied Kalman filtering, 2nd edn. Robert Grover Brown and Patrick Y. C. Hwang, Wiley, New York, 1992. ISBN 0‐471‐52573‐1, 512 pp., $62.95. , 1992 .

[17]  Sofia C. Olhede,et al.  On the Analytic Wavelet Transform , 2007, IEEE Transactions on Information Theory.

[18]  Richard G. Baraniuk,et al.  Multiple Window Time Varying Spectrum Estimation , 2000 .

[19]  A M Amjad,et al.  A framework for the analysis of mixed time series/point process data--theory and application to the study of physiological tremor, single motor unit discharges and electromyograms. , 1995, Progress in biophysics and molecular biology.

[20]  Sofia C. Olhede,et al.  Generalized Morse Wavelets as a Superfamily of Analytic Wavelets , 2012, IEEE Transactions on Signal Processing.

[21]  Laura Astolfi,et al.  A new Kalman filter approach for the estimation of high-dimensional time-variant multivariate AR models and its application in analysis of laser-evoked brain potentials , 2010, NeuroImage.

[22]  C. Striebel,et al.  On the maximum likelihood estimates for linear dynamic systems , 1965 .

[23]  S. Haykin,et al.  Multiple window time-frequency distribution and coherence of EEG using Slepian sequences and Hermite functions , 1999, IEEE Transactions on Biomedical Engineering.

[24]  A. Bruns.  Fourier-, Hilbert- and wavelet-based signal analysis: are they really different approaches? , 2004, Journal of Neuroscience Methods.

[25]  A. Engel,et al.  Spectral fingerprints of large-scale neuronal interactions , 2012, Nature Reviews Neuroscience.

[26]  John-Stuart Brittain,et al.  Optimal spectral tracking—Adapting to dynamic regime change , 2011, Journal of Neuroscience Methods.

[27]  J. Tukey,et al.  Variations of Box Plots , 1978 .

[28]  P. Welch.  The use of fast Fourier transform for the estimation of power spectra: A method based on time averaging over short, modified periodograms , 1967 .

[29]  D. Thomson,et al.  Spectrum estimation and harmonic analysis , 1982, Proceedings of the IEEE.

[30]  John-Stuart Brittain,et al.  Optimal spectral tracking—With application to speed dependent neural modulation of tibialis anterior during human treadmill walking , 2009, Journal of Neuroscience Methods.

[31]  T. Başar,et al.  A New Approach to Linear Filtering and Prediction Problems , 2001 .

[32]  Andrew T. Walden,et al.  Wavelet Coherence for Certain Nonstationary Bivariate Processes , 2011, IEEE Transactions on Signal Processing.

[33]  D. B. Preston.  Spectral Analysis and Time Series , 1983 .

[34]  Donald B. Percival,et al.  Spectral Analysis for Physical Applications , 1993 .

[35]  Sofia C. Olhede,et al.  Generalized Morse wavelets , 2002, IEEE Trans. Signal Process..

[36]  Steven G. Johnson,et al.  The Design and Implementation of FFTW3 , 2005, Proceedings of the IEEE.

[37]  G. Carter,et al.  Estimation of the magnitude-squared coherence function via overlapped fast Fourier transform processing , 1973 .

[38]  J. Schoffelen,et al.  Comparing spectra and coherences for groups of unequal size , 2007, Journal of Neuroscience Methods.

[39]  G. Carter.  Coherence and time delay estimation , 1987, Proceedings of the IEEE.

[40]  Andrew T. Walden,et al.  Statistical Properties for Coherence Estimators From Evolutionary Spectra , 2012, IEEE Transactions on Signal Processing.

[41]  John-Stuart Brittain,et al.  Single-Trial Multiwavelet Coherence in Application to Neurophysiological Time Series , 2007, IEEE Transactions on Biomedical Engineering.

[42]  D. Brillinger.  Fourier analysis of stationary processes , 1974 .

[43]  Dirk Ostwald,et al.  A tutorial on variational Bayes for latent linear stochastic time-series models , 2014 .

[44]  D. Gabor,et al.  Theory of communication. Part 1: The analysis of information , 1946 .

[45]  Andrew T. Walden,et al.  A Statistical Analysis of Morse Wavelet Coherence , 2010, IEEE Transactions on Signal Processing.

[46]  Florian Nadel,et al.  Stochastic Processes And Filtering Theory , 2016 .

[47]  D. Brillinger.  Time series - data analysis and theory , 1981, Classics in applied mathematics.

[48]  Jianfeng Feng,et al.  Detecting time-dependent coherence between non-stationary electrophysiological signals—A combined statistical and time–frequency approach , 2006, Journal of Neuroscience Methods.