Method for exploratory cluster analysis and visualisation of single-trial ERP ensembles

BACKGROUND The validity of ensemble averaging on event-related potential (ERP) data has been questioned, due to its assumption that the ERP is identical across trials. Thus, there is a need for preliminary testing for cluster structure in the data. NEW METHOD We propose a complete pipeline for the cluster analysis of ERP data. To increase the signal-to-noise (SNR) ratio of the raw single-trials, we used a denoising method based on Empirical Mode Decomposition (EMD). Next, we used a bootstrap-based method to determine the number of clusters, through a measure called the Stability Index (SI). We then used a clustering algorithm based on a Genetic Algorithm (GA) to define initial cluster centroids for subsequent k-means clustering. Finally, we visualised the clustering results through a scheme based on Principal Component Analysis (PCA). RESULTS After validating the pipeline on simulated data, we tested it on data from two experiments - a P300 speller paradigm on a single subject and a language processing study on 25 subjects. Results revealed evidence for the existence of 6 clusters in one experimental condition from the language processing study. Further, a two-way chi-square test revealed an influence of subject on cluster membership. COMPARISON WITH EXISTING METHOD(S) Our analysis operates on denoised single-trials, the number of clusters are determined in a principled manner and the results are presented through an intuitive visualisation. CONCLUSIONS Given the cluster structure in some experimental conditions, we suggest application of cluster analysis as a preliminary step before ensemble averaging.

[1]  J Anderson,et al.  Classification of single-trial ERP sub-types: application of globally optimal vector quantization using simulated annealing. , 1995, Electroencephalography and clinical neurophysiology.

[2]  Wei Qiu,et al.  Adaptive filtering of evoked potentials with radial-basis-function neural network prefilter , 2002, IEEE Transactions on Biomedical Engineering.

[3]  J. Michael Herrmann,et al.  Denoising and Averaging Techniques for Electrophysiological Data , 2009 .

[4]  Tom Eichele,et al.  Tracking pattern learning with single-trial event-related potentials , 2006, Clinical Neurophysiology.

[5]  J. A. Hartigan,et al.  A k-means clustering algorithm , 1979 .

[6]  C. A. Murthy,et al.  Genetic Algorithm with Elitist Model and Its Convergence , 1996, Int. J. Pattern Recognit. Artif. Intell..

[7]  S. Tseng,et al.  Evaluation of parametric methods in EEG signal analysis. , 1995, Medical engineering & physics.

[8]  E. Basar,et al.  Principles of oscillatory brain dynamics and a treatise of recognition of faces and facial expressions. , 2006, Progress in brain research.

[9]  Goldberg,et al.  Genetic algorithms , 1993, Robust Control Systems with Genetic Algorithms.

[10]  Klaus-Robert Müller,et al.  The BCI competition 2003: progress and perspectives in detection and discrimination of EEG single trials , 2004, IEEE Transactions on Biomedical Engineering.

[11]  Gabriel Rilling,et al.  On empirical mode decomposition and its algorithms , 2003 .

[12]  John W. Sammon,et al.  A Nonlinear Mapping for Data Structure Analysis , 1969, IEEE Transactions on Computers.

[13]  Steven Kay,et al.  Modern Spectral Estimation: Theory and Application , 1988 .

[14]  N. Huang,et al.  The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis , 1998, Proceedings of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences.

[15]  Douglas Saddy,et al.  Processing polarity items: Contrastive licensing costs , 2004, Brain and Language.

[16]  D. Liberati,et al.  A parametric method of identification of single-trial event-related potentials in the brain , 1988, IEEE Transactions on Biomedical Engineering.

[17]  Sungho Kim,et al.  Robust Object Categorization and Segmentation Motivated by Visual Contexts in the Human Visual System , 2011, EURASIP J. Adv. Signal Process..

[18]  Isabelle Guyon,et al.  A Stability Based Method for Discovering Structure in Clustered Data , 2001, Pacific Symposium on Biocomputing.

[19]  R. Frisch.  Correlation and scatter in statistical variables , 1951 .

[20]  Peter beim Graben,et al.  Diagnosis and repair of negative polarity constructions in the light of symbolic resonance analysis , 2006, Brain and Language.

[21]  Ujjwal Maulik,et al.  Genetic algorithm-based clustering technique , 2000, Pattern Recognit..

[22]  Philip Chan,et al.  Determining the number of clusters/segments in hierarchical clustering/segmentation algorithms , 2004, 16th IEEE International Conference on Tools with Artificial Intelligence.

[23]  R. Quian Quiroga,et al.  Single-trial event-related potentials with wavelet denoising , 2003, Clinical Neurophysiology.

[24]  J. Dunn.  Well-Separated Clusters and Optimal Fuzzy Partitions , 1974 .

[25]  Slawomir J. Nasuto,et al.  Evaluation of Empirical Mode Decomposition for Event-Related Potential Analysis , 2011, EURASIP J. Adv. Signal Process..

[26]  David Zipser,et al.  Feature Discovery by Competive Learning , 1986, Cogn. Sci..

[27]  Johannes D. Veldhuis,et al.  NPY neurons as a critical hypothalamic node for the control of GH release relative to food intake , 2014 .

[28]  N V Thakor,et al.  Adaptive Filterng of Evoked Potentials , 1987, IEEE Transactions on Biomedical Engineering.

[29]  B.H. Jansen,et al.  A fuzzy clustering approach to EP estimation , 1997, IEEE Transactions on Biomedical Engineering.

[30]  Azeddine Beghdadi,et al.  Noise filtering using Empirical Mode Decomposition , 2007, 2007 9th International Symposium on Signal Processing and Its Applications.

[31]  Gabriel Rilling,et al.  Detrending and denoising with empirical mode decompositions , 2004, 2004 12th European Signal Processing Conference.

[32]  O. Jensen,et al.  Posterior α activity is not phase-reset by visual stimuli , 2006 .

[33]  Dan Boneh,et al.  On genetic algorithms , 1995, COLT '95.

[34]  Todd C. Handy,et al.  Event-related potentials : a methods handbook , 2005 .

[35]  E. Donchin,et al.  Talking off the top of your head: toward a mental prosthesis utilizing event-related brain potentials. , 1988, Electroencephalography and clinical neurophysiology.

[36]  B H Jansen,et al.  Selective Averaging of Evoked Potentials using Trajectory-Based Clustering , 1994, Methods of Information in Medicine.

[37]  Vipin Kumar,et al.  Introduction to Data Mining , 2022, Data Mining and Machine Learning Applications.

[38]  O. Jensen,et al.  Posterior alpha activity is not phase-reset by visual stimuli. , 2006, Proceedings of the National Academy of Sciences of the United States of America.

[39]  A Bezerianos,et al.  Robust moving averages, with Hopfield neural network implementation, for monitoring evoked potential signals. , 1997, Electroencephalography and clinical neurophysiology.

[40]  Robert Tibshirani,et al.  Estimating the number of clusters in a data set via the gap statistic , 2000 .

[41]  Hava T. Siegelmann,et al.  Overcoming selective ensemble averaging: unsupervised identification of event-related brain potentials , 2000, IEEE Transactions on Biomedical Engineering.

[42]  Terrence J. Sejnowski,et al.  Independent Component Analysis Using an Extended Infomax Algorithm for Mixed Subgaussian and Supergaussian Sources , 1999, Neural Computation.

[43]  Giuseppe Baselli,et al.  An adaptive neuro-fuzzy method (ANFIS) for estimating single-trial movement-related potentials , 2004, Biological Cybernetics.

[44]  Peter A Tass,et al.  Stochastic phase resetting of stimulus-locked responses of two coupled oscillators: transient response clustering, synchronization, and desynchronization. , 2003, Chaos.

[45]  Jeanny Hérault,et al.  Curvilinear component analysis: a self-organizing neural network for nonlinear mapping of data sets , 1997, IEEE Trans. Neural Networks.