EEG Signal Processing and Machine Learning

✍ Scribed by Saeid Sanei, Jonathon A. Chambers

• Publisher: Wiley
• Year: 2021
• Tongue: English
• Leaves: 751
• Edition: 2
• Category: Library

✦ Synopsis

Explore cutting edge techniques at the forefront of electroencephalogram research and artificial intelligence from leading voices in the field

The newly revised Second Edition of EEG Signal Processing and Machine Learning delivers an inclusive and thorough exploration of new techniques and outcomes in electroencephalogram (EEG) research in the areas of analysis, processing, and decision making about a variety of brain states, abnormalities, and disorders using advanced signal processing and machine learning techniques. The book content is substantially increased upon that of the first edition and, while it retains what made the first edition so popular, is composed of more than 50% new material.

The distinguished authors have included new material on tensors for EEG analysis and sensor fusion, as well as new chapters on mental fatigue, sleep, seizure, neurodevelopmental diseases, BCI, and psychiatric abnormalities. In addition to including a comprehensive chapter on machine learning, machine learning applications have been added to almost all the chapters. Moreover, multimodal brain screening, such as EEG-fMRI, and brain connectivity have been included as two new chapters in this new edition.

Readers will also benefit from the inclusion of:

A thorough introduction to EEGs, including neural activities, action potentials, EEG generation, brain rhythms, and EEG recording and measurement

An exploration of brain waves, including their generation, recording, and instrumentation, including abnormal EEG patterns and the effects of ageing and mental disorders

A treatment of mathematical models for normal and abnormal EEGs

Discussions of the fundamentals of EEG signal processing, including statistical properties, linear and nonlinear systems, frequency domain approaches, tensor factorization, diffusion adaptive filtering, deep neural networks, and complex-valued signal processing

Perfect for biomedical engineers, neuroscientists, neurophysiologists, psychiatrists, engineers, and students and researchers in the above areas, the Second Edition of EEG Signal Processing and Machine Learning will also earn a place in the libraries of undergraduate and postgraduate Biomedical Engineering and Neuroscience, including Epileptology, students.

✦ Table of Contents

Cover
Title Page
Copyright Page
Contents
Preface to the Second Edition
Preface to the First Edition
List of Abbreviations
Chapter 1 Introduction to Electroencephalography
1.1 Introduction
1.2 History
1.3 Neural Activities
1.4 Action Potentials
1.5 EEG Generation
1.6 The Brain as a Network
1.7 Summary
References
Chapter 2 EEG Waveforms
2.1 Brain Rhythms
2.2 EEG Recording and Measurement
2.2.1 Conventional Electrode Positioning
2.2.2 Unconventional and Special Purpose EEG Recording Systems
2.2.3 Invasive Recording of Brain Potentials
2.2.4 Conditioning the Signals
2.3 Sleep
2.4 Mental Fatigue
2.5 Emotions
2.6 Neurodevelopmental Disorders
2.7 Abnormal EEG Patterns
2.8 Ageing
2.9 Mental Disorders
2.9.1 Dementia
2.9.2 Epileptic Seizure and Nonepileptic Attacks
2.9.3 Psychiatric Disorders
2.9.4 External Effects
2.10 Summary
References
Chapter 3 EEG Signal Modelling
3.1 Introduction
3.2 Physiological Modelling of EEG Generation
3.2.1 Integrate-and-Fire Models
3.2.2 Phase-Coupled Models
3.2.3 Hodgkin–Huxley Model
3.2.4 Morris–Lecar Model
3.3 Generating EEG Signals Based on Modelling the Neuronal Activities
3.4 Mathematical Models Derived Directly from the EEG Signals
3.4.1 Linear Models
3.4.1.1 Prediction Method
3.4.1.2 Prony's Method
3.4.2 Nonlinear Modelling
3.4.3 Gaussian Mixture Model
3.5 Electronic Models
3.5.1 Models Describing the Function of the Membrane
3.5.1.1 Lewis Membrane Model
3.5.1.2 Roy Membrane Model
3.5.2 Models Describing the Function of a Neuron
3.5.2.1 Lewis Neuron Model
3.5.2.2 The Harmon Neuron Model
3.5.3 A Model Describing the Propagation of the Action Pulse in an Axon
3.5.4 Integrated Circuit Realizations
3.6 Dynamic Modelling of Neuron Action Potential Threshold
3.7 Summary
References
Chapter 4 Fundamentals of EEG Signal Processing
4.1 Introduction
4.2 Nonlinearity of the Medium
4.3 Nonstationarity
4.4 Signal Segmentation
4.5 Signal Transforms and Joint Time–Frequency Analysis
4.5.1 Wavelet Transform
4.5.1.1 Continuous Wavelet Transform
4.5.1.2 Examples of Continuous Wavelets
4.5.1.3 Discrete-Time Wavelet Transform
4.5.1.4 Multiresolution Analysis
4.5.1.5 Wavelet Transform Using Fourier Transform
4.5.1.6 Reconstruction
4.5.2 Synchro-Squeezed Wavelet Transform
4.5.3 Ambiguity Function and the Wigner–Ville Distribution
4.6 Empirical Mode Decomposition
4.7 Coherency, Multivariate Autoregressive Modelling, and Directed Transfer Function
4.8 Filtering and Denoising
4.9 Principal Component Analysis
4.9.1 Singular Value Decomposition
4.10 Summary
References
Chapter 5 EEG Signal Decomposition
5.1 Introduction
5.2 Singular Spectrum Analysis
5.2.1 Decomposition
5.2.2 Reconstruction
5.3 Multichannel EEG Decomposition
5.3.1 Independent Component Analysis
5.3.2 Instantaneous BSS
5.3.3 Convolutive BSS
5.3.3.1 General Applications
5.3.3.2 Application of Convolutive BSS to EEG
5.4 Sparse Component Analysis
5.4.1 Standard Algorithms for Sparse Source Recovery
5.4.1.1 Greedy-Based Solution
5.4.1.2 Relaxation-Based Solution
5.4.2 k-Sparse Mixtures
5.5 Nonlinear BSS
5.6 Constrained BSS
5.7 Application of Constrained BSS; Example
5.8 Multiway EEG Decompositions
5.8.1 Tensor Factorization for BSS
5.8.2 Solving BSS of Nonstationary Sources Using Tensor Factorization
5.9 Tensor Factorization for Underdetermined Source Separation
5.10 Tensor Factorization for Separation of Convolutive Mixtures in the Time Domain
5.11 Separation of Correlated Sources via Tensor Factorization
5.12 Common Component Analysis
5.13 Canonical Correlation Analysis
5.14 Summary
References
Chapter 6 Chaos and Dynamical Analysis
6.1 Introduction to Chaos and Dynamical Systems
6.2 Entropy
6.3 Kolmogorov Entropy
6.4 Multiscale Fluctuation-Based Dispersion Entropy
6.5 Lyapunov Exponents
6.6 Plotting the Attractor Dimensions from Time Series
6.7 Estimation of Lyapunov Exponents from Time Series
6.7.1 Optimum Time Delay
6.7.2 Optimum Embedding Dimension
6.8 Approximate Entropy
6.9 Using Prediction Order
6.10 Summary
References
Chapter 7 Machine Learning for EEG Analysis
7.1 Introduction
7.2 Clustering Approaches
7.2.1 k-Means Clustering Algorithm
7.2.2 Iterative Self-Organizing Data Analysis Technique
7.2.3 Gap Statistics
7.2.4 Density-Based Clustering
7.2.5 Affinity-Based Clustering
7.2.6 Deep Clustering
7.2.7 Semi-Supervised Clustering
7.2.7.1 Basic Semi-Supervised Techniques
7.2.7.2 Deep Semi-Supervised Techniques
7.2.8 Fuzzy Clustering
7.3 Classification Algorithms
7.3.1 Decision Trees
7.3.2 Random Forest
7.3.3 Linear Discriminant Analysis
7.3.4 Support Vector Machines
7.3.5 k-Nearest Neighbour
7.3.6 Gaussian Mixture Model
7.3.7 Logistic Regression
7.3.8 Reinforcement Learning
7.3.9 Artificial Neural Networks
7.3.9.1 Deep Neural Networks
7.3.9.2 Convolutional Neural Networks
7.3.9.3 Autoencoders
7.3.9.4 Variational Autoencoder
7.3.9.5 Recent DNN Approaches
7.3.9.6 Spike Neural Networks
7.3.9.7 Applications of DNNs to EEG
7.3.10 Gaussian Processes
7.3.11 Neural Processes
7.3.12 Graph Convolutional Networks
7.3.13 Naïve Bayes Classifier
7.3.14 Hidden Markov Model
7.3.14.1 Forward Algorithm
7.3.14.2 Backward Algorithm
7.3.14.3 HMM Design
7.4 Common Spatial Patterns
7.5 Summary
References
Chapter 8 Brain Connectivity and Its Applications
8.1 Introduction
8.2 Connectivity through Coherency
8.3 Phase-Slope Index
8.4 Multivariate Directionality Estimation
8.4.1 Directed Transfer Function
8.4.2 Direct DTF
8.4.3 Partial Directed Coherence
8.5 Modelling the Connectivity by Structural Equation Modelling
8.6 Stockwell Time–Frequency Transform for Connectivity Estimation
8.7 Inter-Subject EEG Connectivity
8.7.1 Objectives
8.7.2 Technological Relevance
8.8 State-Space Model for Estimation of Cortical Interactions
8.9 Application of Cooperative Adaptive Filters
8.9.1 Use of Cooperative Kalman Filter
8.9.2 Task-Related Adaptive Connectivity
8.9.3 Diffusion Adaptation
8.9.4 Brain Connectivity for Cooperative Adaptation
8.9.5 Other Applications of Cooperative Learning and Brain Connectivity Estimation
8.10 Graph Representation of Brain Connectivity
8.11 Tensor Factorization Approach
8.12 Summary
References
Chapter 9 Event-Related Brain Responses
9.1 Introduction
9.2 ERP Generation and Types
9.2.1 P300 and its Subcomponents
9.3 Detection, Separation, and Classification of P300 Signals
9.3.1 Using ICA
9.3.2 Estimation of Single-Trial Brain Responses by Modelling the ERP Waveforms
9.3.3 ERP Source Tracking in Time
9.3.4 Time–Frequency Domain Analysis
9.3.5 Application of Kalman Filter
9.3.6 Particle Filtering and its Application to ERP Tracking
9.3.7 Variational Bayes Method
9.3.8 Prony's Approach for Detection of P300 Signals
9.3.9 Adaptive Time–Frequency Methods
9.4 Brain Activity Assessment Using ERP
9.5 Application of P300 to BCI
9.6 Summary
References
Chapter 10 Localization of Brain Sources
10.1 Introduction
10.2 General Approaches to Source Localization
10.2.1 Dipole Assumption
10.3 Head Model
10.4 Most Popular Brain Source Localization Approaches
10.4.1 EEG Source Localization Using Independent Component Analysis
10.4.2 MUSIC Algorithm
10.4.3 LORETA Algorithm
10.4.4 FOCUSS Algorithm
10.4.5 Standardized LORETA
10.4.6 Other Weighted Minimum Norm Solutions
10.4.7 Evaluation Indices
10.4.8 Joint ICA–LORETA Approach
10.5 Forward Solutions to the Localization Problem
10.5.1 Partially Constrained BSS Method
10.5.2 Constrained Least-Squares Method for Localization of P3a and P3b
10.5.3 Spatial Notch Filtering Approach
10.6 The Methods Based on Source Tracking
10.6.1 Deflation Beamforming Approach for EEG/MEG Multiple Source Localization
10.6.2 Hybrid Beamforming – Particle Filtering
10.7 Determination of the Number of Sources from the EEG/MEG Signals
10.8 Other Hybrid Methods
10.9 Application of Machine Learning for EEG/MEG Source Localization
10.10 Summary
References
Chapter 11 Epileptic Seizure Prediction, Detection, and Localization
11.1 Introduction
11.2 Seizure Detection
11.2.1 Adult Seizure Detection from EEGs
11.2.2 Detection of Neonatal Seizure
11.3 Chaotic Behaviour of Seizure EEG
11.4 Seizure Detection from Brain Connectivity
11.5 Prediction of Seizure Onset from EEG
11.6 Intracranial and Joint Scalp–Intracranial Recordings for IED Detection
11.6.1 Introduction to IED
11.6.2 iEED-Times IED Detection from Scalp EEG
11.6.3 A Multiview Approach to IED Detection
11.6.4 Coupled Dictionary Learning for IED Detection
11.6.5 A Deep Learning Approach to IED Detection
11.7 Fusion of EEG–fMRI Data for Seizure Prediction
11.8 Summary
References
Chapter 12 Sleep Recognition, Scoring, and Abnormalities
12.1 Introduction
12.1.1 Definition of Sleep
12.1.2 Sleep Disorder
12.2 Stages of Sleep
12.2.1 NREM Sleep
12.2.2 REM Sleep
12.3 The Influence of Circadian Rhythms
12.4 Sleep Deprivation
12.5 Psychological Effects
12.6 EEG Sleep Analysis and Scoring
12.6.1 Detection of the Rhythmic Waveforms and Spindles Employing Blind Source Separation
12.6.2 Time–Frequency Analysis of Sleep EEG Using Matching Pursuit
12.6.3 Detection of Normal Rhythms and Spindles Using Higher-Order Statistics
12.6.4 Sleep Scoring Using Tensor Factorization
12.6.5 Application of Neural Networks
12.6.6 Model-Based Analysis
12.7 Detection and Monitoring of Brain Abnormalities during Sleep by EEG and Multimodal PSG Analysis
12.7.1 Analysis of Sleep Apnoea
12.7.2 EEG and Fibromyalgia Syndrome
12.7.3 Sleep Disorders of Neonates
12.8 Dreams and Nightmares
12.9 EEG and Consciousness
12.10 Functional Brain Connectivity for Sleep Analysis
12.11 Summary
References
Chapter 13 EEG-Based Mental Fatigue Monitoring
13.1 Introduction
13.2 Feature-Based Machine Learning Approaches
13.2.1 Hidden Markov Model Application
13.2.2 Kernel Principal Component Analysis and Hidden Markov Model
13.2.3 Regression-Based Fatigue Estimation
13.2.4 Regularized Regression
13.2.5 Other Feature-Based Approaches
13.3 Measurement of Brain Synchronization and Coherency
13.3.1 Linear Measure of Synchronization
13.3.2 Nonlinear Measure of Synchronization
13.4 Evaluation of ERP for Mental Fatigue
13.5 Separation of P3a and P3b
13.6 A Hybrid EEG–ERP-Based Method for Fatigue Analysis Using an Auditory Paradigm
13.7 Assessing Mental Fatigue by Measuring Functional Connectivity
13.8 Deep Learning Approaches for Fatigue Evaluation
13.9 Summary
References
Chapter 14 EEG-Based Emotion Recognition and Classification
14.1 Introduction
14.1.1 Theories and Emotion Classification
14.1.2 The Physiological Effects of Emotions
14.1.3 Psychology and Psychophysiology of Emotion
14.1.4 Emotion Regulation
14.1.4.1 Agency and Intentionality
14.1.4.2 Norm Violation
14.1.4.3 Guilt
14.1.4.4 Shame
14.1.4.5 Embarrassment
14.1.4.6 Pride
14.1.4.7 Indignation and Anger
14.1.4.8 Contempt
14.1.4.9 Pity and Compassion
14.1.4.10 Awe and Elevation
14.1.4.11 Gratitude
14.1.5 Emotion-Provoking Stimuli
14.2 Effect of Emotion on the Brain
14.2.1 ERP Change Due to Emotion
14.2.2 Changes of Normal Brain Rhythms with Emotion
14.2.3 Emotion and Lateral Brain Engagement
14.2.4 Perception of Odours and Emotion: Why Are They Related?
14.3 Emotion-Related Brain Signal Processing and Machine Learning
14.3.1 Evaluation of Emotion Based on the Changes in Brain Rhythms
14.3.2 Brain Asymmetricity and Connectivity for Emotion Evaluation
14.3.3 Changes in ERPs for Emotion Recognition
14.3.4 Combined Features for Emotion Analysis
14.4 Other Physiological Measurement Modalities Used for Emotion Study
14.5 Applications
14.6 Pain Assessment Using EEG
14.7 Emotion Elicitation and Induction through Virtual Reality
14.8 Summary
References
Chapter 15 EEG Analysis of Neurodegenerative Diseases
15.1 Introduction
15.2 Alzheimer's Disease
15.2.1 Application of Brain Connectivity Estimation to AD and MCI
15.2.2 ERP-Based AD Monitoring
15.2.3 Other Approaches to EEG-Based AD Monitoring
15.3 Motor Neuron Disease
15.4 Parkinson's Disease
15.5 Huntington's Disease
15.6 Prion Disease
15.7 Behaviour Variant Frontotemporal Dementia
15.8 Lewy Body Dementia
15.9 Summary
References
Chapter 16 EEG As A Biomarker for Psychiatric and Neurodevelopmental Disorders
16.1 Introduction
16.1.1 History
16.1.1.1 Different Psychiatric and Neurodevelopmental Disorders
16.1.1.2 NDD Diagnosis
16.2 EEG Analysis for Different NDDs
16.2.1 ADHD
16.2.1.1 ADHD Symptoms and Possible Treatment
16.2.1.2 EEG-Based Diagnosis of ADHD
16.2.2 ASD
16.2.2.1 ASD Symptoms and Possible Treatment
16.2.2.2 EEG-Based Diagnosis of ASD
16.2.3 Mood Disorder
16.2.3.1 EEG for Monitoring Depression
16.2.3.2 EEG for Monitoring Bipolar Disorder
16.2.4 Schizophrenia
16.2.4.1 Schizophrenia Symptoms and Management
16.2.4.2 EEG as the Biomarker for Schizophrenia
16.2.5 Anxiety (and Panic) Disorder
16.2.5.1 Definition and Symptoms
16.2.5.2 EEG for Assessing Anxiety
16.2.6 Insomnia
16.2.6.1 Symptoms of Insomnia
16.2.6.2 EEG for Insomnia Analysis
16.2.7 Schizotypal Personality Disorder
16.2.7.1 What Is Schizotypal Disorder?
16.2.7.2 EEG Manifestation of Schizotypal
16.3 Summary
References
Chapter 17 Brain–Computer Interfacing Using EEG
17.1 Introduction
17.1.1 State of the Art in BCI
17.1.2 BCI Terms and Definitions
17.1.3 Popular BCI Directions
17.1.4 Virtual Environment for BCI
17.1.5 Evolution of BCI Design
17.2 BCI-Related EEG Components
17.2.1 Readiness Potential and Its Detection
17.2.2 ERD and ERS
17.2.3 Transient Beta Activity after the Movement
17.2.4 Gamma Band Oscillations
17.2.5 Long Delta Activity
17.2.6 ERPs
17.3 Major Problems in BCI
17.3.1 Preprocessing of the EEGs
17.4 Multidimensional EEG Decomposition
17.4.1 Space–Time–Frequency Method
17.4.2 Parallel Factor Analysis
17.5 Detection and Separation of ERP Signals
17.6 Estimation of Cortical Connectivity
17.7 Application of Common Spatial Patterns
17.8 Multiclass Brain–Computer Interfacing
17.9 Cell-Cultured BCI
17.10 Recent BCI Applications
17.11 Neurotechnology for BCI
17.12 Joint EEG and Other Brain-Scanning Modalities for BCI
17.12.1 Joint EEG–fNIRS for BCI
17.12.2 Joint EEG–MEG for BCI
17.13 Performance Measures for BCI Systems
17.14 Summary
References
Chapter 18 Joint Analysis of EEG and Other Simultaneously Recorded Brain Functional Neuroimaging Modalities
18.1 Introduction
18.2 Fundamental Concepts
18.2.1 Functional Magnetic Resonance Imaging
18.2.1.1 Blood Oxygenation Level Dependence
18.2.1.2 Popular fMRI Data Formats
18.2.1.3 Preprocessing of fMRI Data
18.2.2 Functional Near-Infrared Spectroscopy
18.2.3 Magnetoencephalography
18.3 Joint EEG–fMRI
18.3.1 Relation Between EEG and fMRI
18.3.2 Model-Based Method for BOLD Detection
18.3.3 Simultaneous EEG–fMRI Recording: Artefact Removal from EEG
18.3.3.1 Gradient Artefact Removal from EEG
18.3.3.2 Ballistocardiogram Artefact Removal from EEG
18.3.4 BOLD Detection in fMRI
18.3.4.1 Implementation of Different NMF Algorithms for BOLD Detection
18.3.4.2 BOLD Detection Experiments
18.3.5 Fusion of EEG and fMRI
18.3.5.1 Extraction of fMRI Time Course from EEG
18.3.5.2 Fusion of EEG and fMRI; Blind Approach
18.3.5.3 Fusion of EEG and fMRI; Model-Based Approach
18.3.6 Application to Seizure Detection
18.3.7 Investigation of Decision Making in the Brain
18.3.8 Application to Schizophrenia
18.3.9 Other Applications
18.4 EEG–NIRS Joint Recording and Fusion
18.5 MEG–EEG Fusion
18.6 Summary
References
Index
EULA