EEG Signal Processing

Description

It begins with an introductory chapter discussing the significance of EEG signal analysis and processing and provides some simple examples.  A useful theoretical and mathematical background for analysis and processing of the EEG signals is developed within the next chapter and the mathematical tools to be applied in the rest of the book are covered. 

 

Impressions from the EEG are next discussed, including normal and abnormal EEGs and the neurological symptoms diagnosed.  The representations of the EEG’s are illustrated, showing EEGs in various domains together with two-dimensional maps of the brain.  Theoretical approaches in EEG modeling are reviewed, such as restoration, enhancement, segmentation, and the removal of different internal and external artifacts from the EEG and ERP signals.  The following chapters review seizure detection and prediction, including chaotic behavior of the EEG sources.  Symptoms for a number of well-known illnesses such as dementia, schizophrenia, and Alzheimer’s diseases found from the EEG and ERPs are explained. The concluding chapters review the monitoring and diagnosis of stages of psychiatric disorders using noninvasive techniques such as from the EEG/ERP, and a number of high potential future topics for research in the area of EEG signal processing are discussed.

Table of Contents

Preface.

List of Abbreviations.

List of Symbols.

1 Introduction to EEG.

1.1 History.

1.2 Neural Activities.

1.3 Action Potentials.

1.4 EEG Generation.

1.5 Brain Rhythms.

1.6 EEG Recording and Measurement.

1.6.1 Conventional Electrode Positioning.

1.6.2 Conditioning the Signals.

1.7 Abnormal EEG Patterns.

1.8 Ageing.

1.9 Mental Disorders.

1.9.1 Dementia.

1.9.2 Epileptic Seizure and Nonepileptic Attacks.

1.9.3 Psychiatric Disorders.

1.9.4 External Effects.

1.10 Summary and Conclusions.

References.

2 Fundamentals of EEG Signal Processing.

2.1 EEG Signal Modelling.

2.1.1 Linear Models.

2.1.2 Nonlinear Modelling.

2.1.3 Generating EEG Signals Based on Modelling the Neuronal Activities.

2.2 Nonlinearity of the Medium.

2.3 Nonstationarity.

2.4 Signal Segmentation.

2.5 Signal Transforms and Joint Time–Frequency Analysis.

2.5.1 Wavelet Transform.

2.5.2 Ambiguity Function and the Wigner–Ville Distribution.

2.6 Coherency, Multivariate Autoregressive (MVAR) Modelling, and Directed Transfer Function (DTF).

2.7 Chaos and Dynamical Analysis.

2.7.1 Entropy.

2.7.2 Kolmogorov Entropy.

2.7.3 Lyapunov Exponents.

2.7.4 Plotting the Attractor Dimensions from the Time Series.

2.7.5 Estimation of Lyapunov Exponents from the Time Series.

2.7.6 Approximate Entropy.

2.7.7 Using the Prediction Order.

2.8 Filtering and Denoising.

2.9 Principal Component Analysis.

2.9.1 Singular-Value Decomposition.

2.10 Independent Component Analysis.

2.10.1 Instantaneous BSS.

2.10.2 Convolutive BSS.

2.10.3 Sparse Component Analysis.

2.10.4 Nonlinear BSS.

2.10.5 Constrained BSS.

2.11 Application of Constrained BSS: Example.

2.12 Signal Parameter Estimation.

2.13 Classification Algorithms.

2.13.1 Support Vector Machines.

2.13.2 The k-Means Algorithm.

2.14 Matching Pursuits.

2.15 Summary and Conclusions.

References.

3 Event-Related Potentials.

3.1 Detection, Separation, Localization, and Classification of P300 Signals.

3.1.1 Using ICA.

3.1.2 Estimating Single Brain Potential Components by Modelling ERP Waveforms.

3.1.3 Source Tracking.

3.1.4 Localization of the ERP.

3.1.5 Time–Frequency Domain Analysis.

3.1.6 Adaptive Filtering Approach.

3.1.7 Prony’s Approach for Detection of P300 Signals.

3.1.8 Adaptive Time–Frequency Methods.

3.2 Brain Activity Assessment Using ERP.

3.3 Application of P300 to BCI.

3.4 Summary and Conclusions.

References.

4 Seizure Signal Analysis.

4.1 Seizure Detection.

4.1.1 Adult Seizure Detection.

4.1.2 Detection of Neonate Seizure.

4.2 Chaotic Behaviour of EEG Sources.

4.3 Predictability of Seizure from the EEGs.

4.4 Fusion of EEG–fMRI Data for Seizure Prediction.

4.5 Summary and Conclusions.

References.

5 EEG Source Localization.

5.1 Introduction.

5.1.1 General Approaches to Source Localization.

5.1.2 Dipole Assumption.

5.2 Overview of the Traditional Approaches.

5.2.1 ICA Method.

5.2.2 MUSIC Algorithm.

5.2.3 LORETA Algorithm.

5.2.4 FOCUSS Algorithm.

5.2.5 Standardized LORETA.

5.2.6 Other Weighted Minimum Norm Solutions.

5.2.7 Evaluation Indices.

5.2.8 Joint ICA–LORETA Approach.

5.2.9 Partially Constrained BSS Method.

5.3 Determination of the Number of Sources.

5.4 Summary and Conclusions.

References.

6 Sleep EEG.

6.1 Stages of Sleep.

6.1.1 NREM Sleep.

6.1.2 REM Sleep.

6.2 The Influence of Circadian Rhythms.

6.3 Sleep Deprivation.

6.4 Psychological Effects.

6.5 Detection and Monitoring of Brain Abnormalities During Sleep by EEG Analysis.

6.5.1 Detection of the Rhythmic Waveforms and Spindles Incorporating Blind Source Separation.

6.5.2 Application of Matching Pursuit.

6.5.3 Detection of Normal Rhythms and Spindles using Higher Order Statistics.

6.5.4 Application of Neural Networks.

6.5.5 Model-Based Analysis.

6.5.6 Hybrid Methods.

6.6 Concluding Remarks.

References.

7 Brain–Computer Interfacing.

7.1 State of the Art in BCI.

7.1.1 ERD and ERS.

7.1.2 Transient Beta Activity after the Movement.

7.1.3 Gamma Band Oscillations.

7.1.4 Long Delta Activity.

7.2 Major Problems in BCI.

7.2.1 Preprocessing of the EEGs.

7.3 Multidimensional EEG Decomposition.

7.3.1 Space–Time–Frequency Method.

7.3.2 Parallel Factor Analysis.

7.4 Detection and Separation of ERP Signals.

7.5 Source Localization and Tracking of the Moving Sources within the Brain.

7.6 Multivariant Autoregressive (MVAR) Modelling and Coherency Maps.

7.7 Estimation of Cortical Connectivity.

7.8 Summary and Conclusions.

References.

Index.