97,99 €
Mehr erfahren.
- Herausgeber: John Wiley & Sons
- Kategorie: Wissenschaft und neue Technologien
- Sprache: Englisch
In this book, the field of adaptive learning and processing is extended to arguably one of its most important contexts which is the understanding and analysis of brain signals. No attempt is made to comment on physiological aspects of brain activity; instead, signal processing methods are developed and used to assist clinical findings. Recent developments in detection, estimation and separation of diagnostic cues from different modality neuroimaging systems are discussed.
These include constrained nonlinear signal processing techniques which incorporate sparsity, nonstationarity, multimodal data, and multiway techniques.
Key features:
- Covers advanced and adaptive signal processing techniques for the processing of electroencephalography (EEG) and magneto-encephalography (MEG) signals, and their correlation to the corresponding functional magnetic resonance imaging (fMRI)
- Provides advanced tools for the detection, monitoring, separation, localising and understanding of functional, anatomical, and physiological abnormalities of the brain
- Puts a major emphasis on brain dynamics and how this can be evaluated for the assessment of brain activity in various states such as for brain-computer interfacing emotions and mental fatigue analysis
- Focuses on multimodal and multiway adaptive processing of brain signals, the new direction of brain signal research
Sie lesen das E-Book in den Legimi-Apps auf:
Seitenzahl: 882
Veröffentlichungsjahr: 2013
Ähnliche
Contents
Cover
Title Page
Copyright
Preface
Chapter 1 Brain Signals, Their Generation, Acquisition and Properties
1.1 Introduction
1.2 Historical Review of the Brain
1.3 Neural Activities
1.4 Action Potentials
1.5 EEG Generation
1.6 Brain Rhythms
1.7 EEG Recording and Measurement
1.8 Abnormal EEG Patterns
1.9 Aging
1.10 Mental Disorders
1.11 Memory and Content Retrieval
1.12 MEG Signals and Their Generation
1.13 Conclusions
References
Chapter 2 Fundamentals of EEG Signal Processing
2.1 Introduction
2.2 Nonlinearity of the Medium
2.3 Nonstationarity
2.4 Signal Segmentation
2.5 Other Properties of Brain Signals
2.6 Conclusions
References
Chapter 3 EEG Signal Modelling
3.1 Physiological Modelling of EEG Generation
3.2 Mathematical Models
3.3 Generating EEG Signals Based on Modelling the Neuronal Activities
3.4 Electronic Models
3.5 Dynamic Modelling of the Neuron Action Potential Threshold
3.6 Conclusions
References
Chapter 4 Signal Transforms and Joint Time–Frequency Analysis
4.1 Introduction
4.2 Parametric Spectrum Estimation and Z-Transform
4.3 Time–Frequency Domain Transforms
4.4 Ambiguity Function and the Wigner–Ville Distribution
4.5 Hermite Transform
4.6 Conclusions
References
Chapter 5 Chaos and Dynamical Analysis
5.1 Entropy
5.2 Kolmogorov Entropy
5.3 Lyapunov Exponents
5.4 Plotting the Attractor Dimensions from Time Series
5.5 Estimation of Lyapunov Exponents from Time Series
5.6 Approximate Entropy
5.7 Using Prediction Order
5.8 Conclusions
References
Chapter 6 Classification and Clustering of Brain Signals
6.1 Introduction
6.2 Linear Discriminant Analysis
6.3 Support Vector Machines
6.4 k-Means Algorithm
6.5 Common Spatial Patterns
6.6 Conclusions
References
Chapter 7 Blind and Semi-Blind Source Separation
7.1 Introduction
7.2 Singular Spectrum Analysis
7.3 Independent Component Analysis
7.4 Instantaneous BSS
7.5 Convolutive BSS
7.6 Sparse Component Analysis
7.7 Nonlinear BSS
7.8 Constrained BSS
7.9 Application of Constrained BSS; Example
7.10 Nonstationary BSS
7.11 Tensor Factorization for Underdetermined Source Separation
7.12 Tensor Factorization for Separation of Convolutive Mixtures in the Time Domain
7.13 Separation of Correlated Sources via Tensor Factorization
7.14 Conclusions
References
Chapter 8 Connectivity of Brain Regions
8.1 Introduction
8.2 Connectivity Through Coherency
8.3 Phase-Slope Index
8.4 Multivariate Directionality Estimation
8.5 Modelling the Connectivity by Structural Equation Modelling
8.6 EEG Hyper-Scanning and Inter-Subject Connectivity
8.7 State-Space Model for Estimation of Cortical Interactions
8.8 Application of Adaptive Filters
8.9 Tensor Factorization Approach
8.10 Conclusions
References
Chapter 9 Detection and Tracking of Event-Related Potentials
9.1 ERP Generation and Types
9.2 Detection, Separation, and Classification of P300 Signals
9.3 Brain Activity Assessment Using ERP
9.4 Application of P300 to BCI
9.5 Conclusions
References
Chapter 10 Mental Fatigue
10.1 Introduction
10.2 Measurement of Brain Synchronization and Coherency
10.3 Evaluation of ERP for Mental Fatigue
10.4 Separation of P3a and P3b
10.5 A Hybrid EEG-ERP-Based Method for Fatigue Analysis Using an Auditory Paradigm
10.6 Conclusions
References
Chapter 11 Emotion Encoding, Regulation and Control
11.1 Theories and Emotion Classification
11.2 The Effects of Emotions
11.3 Psychology and Psychophysiology of Emotion
11.4 Emotion Regulation
11.5 Emotion-Provoking Stimuli
11.6 Change in the ERP and Normal Brain Rhythms
11.7 Perception of Odours and Emotion: Why Are They Related?
11.8 Emotion-Related Brain Signal Processing
11.9 Other Neuroimaging Modalities Used for Emotion Study
11.10 Applications
11.11 Conclusions
References
Chapter 12 Sleep and Sleep Apnoea
12.1 Introduction
12.2 Stages of Sleep
12.3 The Influence of Circadian Rhythms
12.4 Sleep Deprivation
12.5 Psychological Effects
12.6 Detection and Monitoring of Brain Abnormalities During Sleep by EEG Analysis
12.7 EEG and Fibromyalgia Syndrome
12.8 Sleep Disorders of Neonates
12.9 Dreams and Nightmares
12.10 Conclusions
References
Chapter 13 Brain–Computer Interfacing
13.1 Introduction
13.2 State of the Art in BCI
13.3 BCI-Related EEG Features
13.4 Major Problems in BCI
13.5 Multidimensional EEG Decomposition
13.6 Detection and Separation of ERP Signals
13.7 Estimation of Cortical Connectivity
13.8 Application of Common Spatial Patterns
13.9 Multiclass Brain–Computer Interfacing
13.10 Cell-Cultured BCI
13.11 Conclusions
References
Chapter 14 EEG and MEG Source Localization
14.1 Introduction
14.2 General Approaches to Source Localization
14.3 Most Popular Brain Source Localization Approaches
14.4 Determination of the Number of Sources from the EEG/MEG Signals
14.5 Conclusions
References
Chapter 15 Seizure and Epilepsy
15.1 Introduction
15.2 Types of Epilepsy
15.3 Seizure Detection
15.4 Chaotic Behaviour of EEG Sources
15.5 Predictability of Seizure from the EEGs
15.6 Fusion of EEG – fMRI Data for Seizure Detection and Prediction
15.7 Conclusions
References
Chapter 16 Joint Analysis of EEG and fMRI
16.1 Fundamental Concepts
16.2 Model-Based Method for BOLD Detection
16.3 Simultaneous EEG-fMRI Recording: Artefact Removal from EEG
16.4 BOLD Detection in fMRI
16.5 Fusion of EEG and fMRI
16.6 Application to Seizure Detection
16.7 Conclusons
References
Index
© 2013 John Wiley & Sons, Ltd
Registered office John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, United Kingdom
For details of our global editorial offices, for customer services and for information about how to apply for permission to reuse the copyright material in this book please see our website at www.wiley.com.
The right of the author to be identified as the author of this work has been asserted in accordance with the Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by the UK Copyright, Designs and Patents Act 1988, without the prior permission of the publisher.
Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books.
Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The publisher is not associated with any product or vendor mentioned in this book.
Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. It is sold on the understanding that the publisher is not engaged in rendering professional services and neither the publisher nor the author shall be liable for damages arising herefrom. If professional advice or other expert assistance is required, the services of a competent professional should be sought.
MATLAB® is a trademark of The MathWorks, Inc. and is used with permission. The MathWorks does not warrant the accuracy of the text or exercises in this book. This book’s use or discussion of MATLAB® software or related products does not constitute endorsement or sponsorship by The MathWorks of a particular pedagogical approach or particular use of the MATLAB® software.
Library of Congress Cataloging-in-Publication Data
Sanei, Saeid. Adaptive processing of brain signals / Dr. Saeid Sanei. pages cm Includes bibliographical references and index. ISBN 978-0-470-68613-3 (hardback) 1. Neural networks (Neurobiology). 2. Brain–Physiology. 3. Signal processing–Digital techniques. I. Title.
QP363.3.S26 2013 573.8’5–dc23
2013005190
A catalogue record for this book is available from the British Library
Preface
Since the book, EEG Signal Processing (Sanei, Chambers 2007) was published, there have been more demands for research and development of signal processing tools and algorithms for the much wider exploration of EEG and other neuroimaging systems. More recent advances in digital signal processing are expected to underpin key aspects of the future progress in biomedical research and technology, particularly on measurements and assessment of brain activity and its response to various stimulations.
Although most of the concepts in multi-channel EEG digital signal processing, particularly the use of adaptive techniques and iterative learning algorithms, have their origin in distinct application areas, such as acoustics, communications engineering, speech and biometrics, together with the processing of other physiological signals; it is shown in this book that new approaches stem from many recent neurological, psychological and clinical neuroscience findings.
As well as some fundamental concepts in both the modelling and processing of brain activities, a number of new signal processing topics are explored, including new definitions and algorithms. Multichannel, multidimensional and multiway signal processing, which cover a more inclusive and dynamic description of brain information, are presented in this book. This extends to multimodal data analysis in the context of simultaneous EEG–fMRI recordings and information fusion.
Motivated by research in the field over more than two decades, techniques specifically related to EEG processing, such as brain source localization, detection and classification of event-related potentials, sleep signal analysis, emotion effects, mental fatigue, brain connectivity, seizure detection and prediction, together with brain–computer interfacing are all explained in detail. A comprehensive illustration of new signal processing results in the form of signals, graphs, images, and tables has been provided for a better understanding of the concepts.
Following the history and basic definitions and physiological brain-related concepts, the first few chapters cover a comprehensive overview of the tools and algorithms currently being used for processing EEG and magnetoencephalography (MEG) signals. There is more emphasis put on algorithms utilised for the analysis of scalp recordings.
Detection and tracking of event-related potentials (ERPs), particularly for single-trial estimation, using recent techniques are provided next. ERPs are the brain responses to audio, visual, and tactile stimuli. Therefore, many neurological and psychiatric brain disorders as well as movement-related abnormalities are diagnosed and monitored using these techniques.
Brain connectivity and its dynamics change significantly for various physiological and neurological states of the brain. A comprehensive overview of the techniques plus new directions in the assessment and exploitation of brain connectivity are provided next.
Seizure and epileptic brain discharges are still investigated by many researchers by looking at both scalp EEG signals and intracranial spike recordings using subdural electrodes. Some very recent methods in seizure prediction are demonstrated. This area of research is later extended to a new multimodal methodology employing simultaneous EEG–fMRI data recording and analysis.
The state of the art in the localization of brain signal sources has been followed by two new inclusive forward methods. In places where the desired source signals can be defined using known waveforms forward models result in accurate localization of the sources. A deflation-based localization approach presented here can localize multiple brain signal sources.
Mental fatigue assessment and analysis from EEG is another new research area covered in this book. Combining signal processing and machine learning algorithms leads to establishing a robust and clinically proven method for detection of the stages of mental fatigue. This includes the changes in both brain responses to stimuli and the changes in brain normal rhythms.
Despite a variety of approaches for evaluating emotions from facial expressions and body biometrics, researchers are focussing on the changes in brain rhythms, mainly in terms of source localization and connectivity, using EEG, MEG, and fMRI neuroimaging techniques. A comprehensive overview of the methods and algorithms developed for this purpose is given.
The book also covers very recent machine learning tools in brain–computer interfacing (BCI). Although, the principles of BCI have not changed significantly over the last five years, a number of extensions to well established algorithms, such as common spatial patterns, have been proposed very recently. These approaches are covered in this book.
Unfortunately, the signals are affected by noise and interferences. Much effort, therefore, has to be made to restore the signals. In some applications, such as joint EEG–fMRI recording and analysis, this problem is significant. In this book artefact removal has been extensively addressed and the results of new algorithms discussed and illustrated.
In the preparation of the content of this book, it is assumed that the reader has a background in the fundamentals of digital signal processing and wishes to focus on the processing of brain signals, particularly EEG and MEG. It is hoped that the concepts covered in each chapter provide a foundation for future research and development in the field.
In conclusion, I wish to stress that there is no attempt to challenge any clinical or diagnostic knowledge. Instead, the tools and algorithms described here can, I believe, enhance the clinically-related information within EEG signals significantly and thereby aid physicians in better diagnosis, treatment and monitoring of brain abnormalities.
Before ending this preface I wish to thank most sincerely my friends and colleagues who reviewed this book for their valuable and very useful comments: Ahmad Reza Hosseini-Yazdi, Jonathan Clark, Tracey Kah Mein Lee, and Clive Cheong-Took. Next, my appreciation to my recent PhD students in the University of Surrey, who contributed to provision of the materials in this book, especially, Delaram Jarchi, Hamid Mohseni, Javier Escodero, Foad Ghaderi, Bahador Makkiabadi, Saideh Ferdowsi, and Kostas Eftaxias.
Last but not least, I appreciate the support and patience of my wife Maryam and my sweethearts Erfan, Ideen and Shaghayegh to whom I dedicate this book.
Saeid SaneiJune 2013
1
Brain Signals, Their Generation, Acquisition and Properties
1.1 Introduction
The brain is the most astonishing and complicated part of the human body and is naturally responsible for controlling all other organs. The neural activity of the human brain starts between the 17th and 23rd weeks of prenatal development. It is believed that from this early stage and throughout life electrical signals generated by the brain represent not only the brain function but also the status of the whole body. This assumption provides the motivation to apply advanced digital signal processing methods to the brain functional data, including electroencephalogram (EEG), magnetoencephalogram (MEG), and functional magnetic resonance image (fMRI) sequences. Although the emphasis in this book is on EEG and MEG, there will be some analysis of simultaneously recorded EEG-fMRI sequences too. Other functional brain information, such as that obtained by near-infrared spectroscopy (NIRS) recently developed for recording movement=related cortical potentials and some under-developing imaging systems, such as ultrawideband or microwave brain imaging are rarely referred to.
Nowhere in this book does the author attempt to comment on the physiological aspects of brain activities. However, there are several issues related to the nature of the original sources, their generation, their actual patterns, and the characteristics of the propagating environment.
Understanding of neuronal functions and neurophysiological properties of the brain, together with the mechanisms underlying the generation of signals and their recordings is, however, vital for those who deal with these signals for detection, diagnosis, and treatment of brain disorders and the related diseases. We begin by providing a brief history of EEG recording.
1.2 Historical Review of the Brain
EEG history goes back to the time when, for the first time, some activity of the brain was recorded or displayed. Carlo Matteucci (1811–1868) and Emil Du Bois-Reymond (1818–1896) were the first people to register the electrical signals emitted from muscle nerves using a galvanometer and establish the concept of neurophysiology [1, 2]. However, the concept of action current, introduced by Hermann Von Helmholz [3], clarified and confirmed the negative variations which occur during muscle contraction.
Richard Caton (1842–1926) a scientist from Liverpool, England, used a galvanometer and placed two electrodes over the scalp of a human subject and thereby first recorded brain activity in the form of electrical signals in 1875. Since then, the concepts of electro-(referring to registration of brain electrical activities) encephal-(referring to emitting the signals from the head) and gram (or graphy), meaning drawing or writing, were combined so that the term EEG was henceforth used to denote electrical neural activity of the brain.
Fritsch (1838–1927) and Hitzig (1838–1907) discovered that the human cerebrum can be electrically stimulated. Vasili Yakovlevich Danilevsky (1852–1939) followed Caton's work and finished his PhD thesis in the investigation of brain physiology in 1877 [4]. In this work he investigated the activity of the brain following electrical stimulation as well as spontaneous electrical activity in the brain of animals.
The cerebral electrical activity observed over the visual cortex of different species of animals was reported by Ernst von Fleischl-Marxow (1845–1891). Napoleon Cybulski (1854–1919) provided EEG evidence of an epileptic seizure in a dog caused by electrical stimulation.
The idea of association of epileptic attacks with abnormal electrical discharges was expressed by Kaufman [5]. Pravidch-Neminsky (1879–1952) a Russian physiologist, recorded the EEG from the brain, termed dura, and the intact skull of a dog in 1912. He observed a 12–14 cycle s–1 rhythm under normal conditions which slowed under asphyxia. He later called it the electrocerebrogram.
The discoverer of the existence of human EEG signals was Hans Berger (1873–1941); see Figure 1.1. He began his study of human EEGs in 1920 [6]. Berger is well known by almost all electroencephalographers. He started working with a string galvanometer in 1910, then migrated to a smaller Edelmann model and, after 1924, to a larger Edelmann model. In 1926, Berger started to use a more powerful Siemens double coil galvanometer (attaining a sensitivity of 130 μV cm–1) [7]. His first report of human EEG recordings of 1–3 min duration on photographic paper was in 1929. In this recording he only used a one-channel bipolar method with fronto-occipital leads. Recording of the EEG became popular in 1924. The first report of 1929 by Berger included the alpha rhythm, as the major component of the EEG signals, as described later in this chapter, and the alpha blocking response.
Figure 1.1 Hans Berger (image from http://www.s9.com/Biography/Berger-Hans)
During the 1930s the first EEG recording of sleep spindles was undertaken by Berger. He then reported the effect of hypoxia on the human brain, the nature of several diffuse and localized brain disorders, and gave an inkling of epileptic discharges [8]. During this time another group, established in Berlin-Buch and led by Kornmüller, provided more precise recording of the EEG [9]. Berger was also interested in cerebral localization and particularly in the localization of brain tumours. He also found some correlation between mental activities and the changes in the EEG signals.
Toennies (1902–1970) from the group in Berlin built the first biological amplifier for recording brain potentials. A differential amplifier for recording EEGs was later produced by the Rockefeller foundation in 1932.
The importance of multichannel recordings and using a large number of electrodes to cover a wider brain region was recognised by Kornmüller [10]. The first EEG work focusing on epileptic manifestation and the first demonstration of epileptic spikes were presented by Fischer and Löwenbach [11–13].
In England, W. Gray Walter became the pioneer of clinical EEG. He discovered the foci of slow brain activity (delta waves), which initiated enormous clinical interest in the diagnosis of brain abnormalities. In Brussels, Fredric Bremer (1892–1982) discovered the influence of afferent signals on the state of vigilance [14].
Research activities related to EEGs started in North America in around 1934. In this year, Hallowell Davis illustrated a good alpha rhythm for himself. A cathode ray oscilloscope was used around this date by the group in St. Louis University in Washington, in the study of peripheral nerve potentials. The work on human EEGs started at Harward in Boston and the University of Iowa in the 1930s. The study of epileptic seizure, developed by Fredric Gibbs, was the major work on EEGs during these years, as the realm of epileptic seizure disorders was the domain of their greatest effectiveness. Epileptology may be divided historically into two periods [15]: before and after the advent of EEG. Gibbs and Lennox applied Fischer's idea and the effect of picrotoxin on the cortical EEG in animal and human epileptology. Berger [16] showed a few examples of paroxysmal EEG discharges in a case of presumed petit mal attacks and during a focal motor seizure in a patient with general paresis. For a couple of decades the EEG work focused on the study of epilepsy.
As the other great pioneers of EEG in North America, Hallowel and Pauline Davis were the earliest investigators of the nature of EEG during human sleep. A. L. Loomis, E. N. Harvey, and G. A. Hobart were the first who mathematically studied the human sleep EEG patterns and the stages of sleep. At McGill University, Jasper studied the related behavioural disorder before he found his niche in basic and clinical epileptology [17].
The American EEG society was founded in 1947 and the first international EEG Congress was held in London, U K, around this time. While the EEG studies in Germany were still limited to Berlin, Japan gained attention by the work of Motokawa, a researcher of EEG rhythms [18]. During these years the neurophysiologists demonstrated the thalamocortical relationship through anatomical methods. This led to the development of the concept of centrencephalic epilepsy [19].
Throughout the 1950s the work on EEGs expanded in many different places. During this time surgical operation to remove the epileptic foci became popular and the book entitled Epilepsy and the Functional Anatomy of the Human Brain (Penfield and Jasper) was published. During this time microelectrodes were invented. They were made of metals such as tungsten, or glass, filled with electrolytes, such as potassium chloride, with diameters of less than 3 μm.
Recordings of deep brain EEG sources of a human were first obtained with implanted intracerebral electrodes by Mayer and Hayne (1948). Invention of intracellular microelectrode technology revolutionarized this method and was used in the spinal cord by Brock et al. in 1952 [20], and in the cortex by Phillips in 1961 [21].
Analysis of EEG signals started during the early days of EEG measurement. Berger assisted by Dietch (1932) applied Fourier analysis to EEG sequences which was rapidly developed during the 1950s. Analysis of sleep disorders with EEGs started its development in the 1950s through the work of Kleitman at the University of Chicago.
In the 1960s analysis of the EEGs of full-term and premature newborns began its development [22]. Investigation of evoked potentials (EPs), especially visual EPs, as commonly used for monitoring mental illnesses, progressed during the 1970s.
MEG signals, on the other hand were first measured by David Cohen, a University of Illinois physicist, in 1968 [23] before the availability of the superconducting quantum interference device (SQUID), using a copper induction coil as the detector. To reduce the magnetic background noise, the measurements were made in a magnetically shielded room. The coil detector was barely sensitive enough, resulting in poor, noisy MEG measurements that were difficult to use. Later, Cohen built a better shielded room at Massachusetts Institute of Technology (MIT), and used one of the first SQUID detectors, just developed by James E. Zimmerman, a researcher at Ford Motor Company [24], to again measure MEG signals [1003]. This time the signals were almost as clear as those of EEG. Subsequently, various types of spontaneous and evoked MEGs began to be measured.
Using a single SQUID detector to successively measure the magnetic field at a number of points around the human head was cumbersome. Therefore, in the 1980s, MEG manufacturers began to fabricate multiple sensors into arrays to cover a larger area of the head. Present-day MEG arrays are set in a helmet-shaped dewar that typically contains 300 sensors, covering most of the head. In this way, MEG signals can be recorded much faster.
One advantage of MEG over EEG signals is their much lower sensitivity to the nonlinearity and non-uniformity in brain tissues. EEG on the other hand, is less noisy and much cheaper than MEG.
The history of EEG and MEG, however, has been a continuous process which started from the early 1800s and has brought daily development of clinical, experimental, and computational studies for discovery, recognition, diagnosis, and treatment of a vast number of neurological and physiological abnormalities of the brain and the rest of the central neural system (CNS) of human beings. Nowadays, EEGs are recorded invasively and non-invasively using fully computerised systems. The EEG machines are equipped with many signal processing tools, delicate and accurate measurement electrodes, and enough memory for very long-term recordings of several hours. Although the MEG machines are expensive due to the technology involved and the requirement for low noise SQUIDs and amplifiers, they are being used as an effective tool for brain source localization. EEG or MEG machines may be integrated with other neuroimaging systems, such as fMRI. Very delicate needle-type electrodes can also be used for recording the EEGs from over the cortex (electrocorticogram), thereby avoiding the attenuation and nonlinearity effects induced by the skull. We next proceed to describe the nature of neural activities within the human brain.
1.3 Neural Activities
The central nervous system (CNS) generally consists of nerve cells and glia cells, which are located between neurons. Each nerve cell consists of axons, dendrites and cell bodies. Nerve cells respond to stimuli and transmit information over long distances. A nerve cell body has a single nucleus, and contains most of the nerve cell metabolism, especially that related to protein synthesis. The proteins created in the cell body are delivered to other parts of the nerve. An axon is a long cylinder, which transmits an electrical impulse and can be several meters long in vertebrates (giraffe axons go from the head to the tip of spine). In humans the length can be a percentage of a millimetre to more than a metre. An axonal transport system for delivering proteins to the ends of the cell exists and the transport system has “molecular motors” which ride upon tubulin rails.
Dendrites are connected to either the axons or dendrites of other cells and receive impulses from other nerves or relay the signals to other nerves. In the human brain each nerve is connected to approximately 10 000 other nerves, mostly through dendritic connections.
The activities in the CNS are mainly related to the synaptic currents transferred between the junctions (called synapses) of axons and dendrites, or dendrites and dendrites of cells. A potential of 60–70 mV with negative polarity may be recorded under the membrane of the cell body. This potential changes with variations in synaptic activities. If an action potential travels along the fibre, which ends in an excitatory synapse, an excitatory following neuron. If two action potentials travel along the same fibre over a short distance, there will be a summation of EPSPs producing an action potential on the postsynaptic neuron providing a certain threshold of membrane potential is reached. If the fibre ends in an inhibitory synapse, then hyperpolarization will occur, indicating an inhibitory postsynaptic potential (IPSP) [25, 26]. Figure 1.2 shows the above activities schematically.
Figure 1.2 The neuron membrane potential changes and current flow during synaptic activation recorded by means of intracellular microelectrodes. Action potentials in the excitatory and inhibitory presynaptic fibre, respectively, lead to EPSP and IPSP in the postsynaptic neuron
Following the generation of an IPSP, there is an overflow of cations from the nerve cell or an inflow of anions into the nerve cell. This flow ultimately causes a change in potential along the nerve cell membrane. Primary transmembranous currents generate secondary inonal currents along the cell membranes in the intra- and extra-cellular space. The portion of these currents that flows through the extracellular space is directly responsible for the generation of field potentials. These field potentials, usually with less than 100 Hz frequency, are called EEGs when there are no changes in the signal average, and called DC potential if there are slow drifts in the average signals, which may mask the actual EEG signals. A combination of EEG and DC potentials is often observed for some abnormalities in the brain, such as seizure (induced by pentylenetetrazol), hypercapnia, and asphyxia [27]. We next focus on the nature of action potentials.
1.4 Action Potentials
The information transmitted by a nerve is called an action potential (AP). APs are caused by an exchange of ions across the neuron membrane and an AP is a temporary change in the membrane potential that is transmitted along the axon. It is usually initiated in the cell body and normally travels in one direction. The membrane potential depolarizes (becomes more positive) producing a spike. After the spike reaches its peak amplitude the membrane repolarizes (becomes more negative). The potential becomes more negative than the resting potential and then returns to normal. The action potentials of most nerves last between 5 and 10 milliseconds.
The conduction velocity of action potentials lies between 1 and 100 m s–1. APs are initiated by many different types of stimuli; sensory nerves respond to many types of stimuli, such as chemical, light, electricity, pressure, touch and stretching. On the other hand, the nerves within the CNS (brain and spinal cord) are mostly stimulated by chemical activity at synapses.
A stimulus must be above a threshold level to set off an AP. Very weak stimuli cause a small local electrical disturbance, but do not produce a transmitted AP. As soon as the stimulus strength goes above the threshold, an action potential appears and travels down the nerve.
The spike of the AP is mainly caused by opening of Na (sodium) channels. The Na pump produces gradients of both Na and K (potassium) ions – both are used to produce the action potential; Na is high outside the cell and low inside. Excitable cells have special Na and K channels with gates that open and close in response to the membrane voltage (voltage-gated channels). Opening the gates of Na channels allows Na to rush into the cell, carrying +Ve charge. This makes the membrane potential positive (depolarization), producing the spike. Figure 1.3 shows the stages of the process during evolution of an action potential for a giant squid. For a human being the amplitude of the AP ranges between approximately −60 and 10 mV. During this process [28]:
Figure 1.3 An action potential (membrane potential) for a giant squid by closing the Na channels and opening K channels. Taken from [28], © John Benjamins Publishing Co.
The nerve requires approximately 2 ms before another stimulus is presented. During this time no AP can be generated. This is called the refractory period. The generation of EEG signals is next described.
1.5 EEG Generation
An EEG signal is a measurement of currents that flow during synaptic excitations of the dendrites of many pyramidal neurons in the cerebral cortex. When brain cells (neurons) are activated, the synaptic currents are produced within the dendrites. These currents generate a magnetic field measurable by EMG machines and a secondary electrical field over the scalp measurable by EEG systems.
Differences in electrical potentials are caused by summed postsynaptic graded potentials from pyramidal cells that create electric dipoles between the soma (body of a neuron) and apical dendrites which branch from neurons (Figure 1.4). The current in the brain is generated mostly due to pumping the positive ions of sodium, Na+, potassium, K+, calcium, or Ca++, and the negative ion of Cl−, through the neuron membranes in the direction governed by the membrane potential [29].
Figure 1.4 Structure of a neuron, adapted from [29]
The human head consists of different layers, including scalp, skull, brain and many other thin layers in between. The head layers have different thickness and current resistivities; the scalp has a thickness of approximately 0.2–0.5 cm and resistivity of 300–400 Ω, the skull has a thickness of 0.3–0.7 cm and resistivity of 10–25 kΩ when measured in vivo. In addition, the scalp consists of different layers, such as skin, connective tissue, which is a thin layer of fat and fibrous tissue lying beneath the skin, the loose areolar connective tissue, and the pericranium which is the periosteum of the skull bones and provides nutrition to bone and capacity for repair. Thebrain is covered by a thin layer called the cortex, which encompasses the entire brain lobes. The cortex has a thickness of 0.1–0.3 cm and an in vivo resistivity of 50–150 Ω. The cortex includes arachnoid, meninges, dura, epidural, and subarachnoid space. The skull attenuates the signals approximately one hundred times more than the soft tissue.
Since the layers have different electrical properties, EEG signals are generally a nonlinear sum of the brain sources. However, since the majority of the sources are cortical, that is, very close to the cortex, this nonlinearity does not significantly affect the common source separation processes. For this reason MEG is more popular for brain source localization.
On the other hand, most of the noise is generated either within the brain (internal noise) or over the scalp (system noise or external noise). Therefore, only large populations of active neurons can generate enough potential to be recordable using the scalp electrodes. These signals are later amplified greatly for display purposes. Approximately 1011 neurons are developed at birth when the CNS becomes complete and functional [30]. This makes an average of 104 neurons per mm3. Neurons are interconnected into neural nets through synapses. Adults have approximately 5.1014 synapses. The number of synapses per neuron increases with age, whereas, the number of neurons decreases with age. From an anatomical point of view the brain may be divided into three parts; the cerebrum, cerebellum, and brain stem (Figure 1.5). The cerebrum consists of both left and right lobes of the brain with highly convoluted surface layers called the cerebral cortex.
Figure 1.5 Diagrammatic representation of the major parts of the brain
The cerebrum includes the regions for movement initiation, conscious awareness of sensation, complex analysis, and expression of emotions and behaviour. The cerebellum coordinates voluntary movements of muscles and maintenance of balance.
The brain stem controls involuntary functions, such as respiration, heart regulation, biorythms, neurohormone and hormone secretion [31].
Based on the above section it is clear that the study of EEGs paves the path for diagnosis of many neurological disorders and other abnormalities in the human body. The acquired EEG signals from a human (and also from animals) may for example be used for investigation of the following clinical problems [31, 32]:
This list confirms the rich potential for EEG analysis and motivates the need for advanced signal processing techniques to aid the clinician in their interpretation. We next proceed to describe the brain rhythms which are expected to be measured within EEG signals.
Some of the mechanisms that generate the EEG signals are known at the cellular level and rest on a balance of excitatory and inhibitory interactions within and between populations of neurons. Although it is well established that the EEG (or MEG) signals result mainly from extracellular current flow, associated with summed postsynaptic potentials in synchronously activated and vertically oriented neurons, the exact neurophysiological mechanisms resulting in such synchronisation to a given frequency band, remain obscure.
1.6 Brain Rhythms
Many brain disorders are diagnosed by visual inspection of EEG signals. Clinical experts in the field are familiar with the manifestation of brain rhythms in the EEG signals. In healthy adults, the amplitudes and frequencies of such signals change from one state of a human to another, such as wakefulness and sleep. The characteristics of the waves also change with age. There are five major brain waves distinguished by their different frequency ranges. These frequency bands from low to high frequencies, respectively, are called alpha (α), theta (θ), beta (β), delta (δ) and gamma (γ). The alpha and beta waves were introduced by Berger in 1929. Jasper and Andrews (1938) used the term “gamma” to refer to the waves of above 30 Hz. The delta rhythm was introduced by Walter (1936) to designate all frequencies below the alpha range. He also introduced theta waves as those having frequencies within the range 4–7.5 Hz. The notion of a theta wave was introduced by Wolter and Dovey in 1944 [33].
Delta waves lie within the range 0.5 to 4 Hz. These waves are primarily associated with deep sleep and may be present in the waking state. It is very easy to confuse artefact signals caused by the large muscles of the neck and jaw with the genuine delta response. This is because the muscles are near the surface of the skin and produce large signals, whereas the signal of interest originates from deep within the brain and is severely attenuated in passing through the skull. Nevertheless, by applying simple signal analysis methods to the EEG, it is very easy to see when the response is caused by excessive movement.
Theta waves lie within the range 4 to 7.5 Hz. The term theta might be chosen to allude to its presumed thalamic origin. Theta waves appear as consciousness slips towards drowsiness. Theta waves have been associated with access to unconscious material, creative inspiration and deep meditation. A theta wave is often accompanied by other frequencies and seems to be related to level of arousal. We know that healers and experienced mediators have an alpha wave which gradually lowers in frequency over long periods of time. The theta wave plays an important role in infancy and childhood. Larger contingents of theta wave activity in the waking adult are abnormal and are caused by various pathological problems. The changes in the rhythm of theta waves are examined for maturational and emotional studies [34].
The alpha waves appear in the posterior half of the head, are usually found over the occipital region of the brain, and can be detected in all parts of brain posterior lobes. For alpha waves the frequency lies within the range 8–13 Hz, and commonly appears as a round or sinusoidal-shaped signal. However, in rare cases it may manifest itself as sharp waves. In such cases, the negative component appears to be sharp and the positive component appears to be rounded, similar to the wave morphology of the rolandic mu (μ) rhythm. Alpha waves have been thought to indicate both a relaxed awareness without any attention or concentration. The alpha wave is the most prominent rhythm in the whole realm of brain activity and possibly covers a greater range than has been previously accepted. You can regularly see a peak in the beta wave range in frequencies even up to 20 Hz, which has the characteristics of an alpha wave state rather those of a beta wave. Again, we very often see a response at 75 Hz which appears in an alpha setting. Most subjects produce some alpha waves with their eyes closed and this is why it has been claimed that it is nothing but a waiting or scanning pattern produced by the visual regions of the brain. It is reduced or eliminated by opening the eyes, by hearing unfamiliar sounds, by anxiety or mental concentration or attention. Albert Einstein could solve complex mathematical problems while remaining in the alpha state; though generally, beta and theta waves are also present. An alpha wave has a higher amplitude over the occipital areas and has an amplitude of normally less than 50 μV. The origin and physiological significance of an alpha wave is still unknown and yet more research has to be undertaken to understand how this phenomenon originates from cortical cells [35].
A beta wave is the electrical activity of the brain varying within the range 14–26 Hz (though in some literature no upper bound is given). A beta wave is the usual waking rhythm of the brain associated with active thinking, active attention, focus on the outside world or solving concrete problems, and is found in normal adults. A high level beta wave may be acquired when a human is in a panic state. Rhythmical beta activity is encountered chiefly over the frontal and central regions. Importantly, a central beta rhythm is related to the rolandic mu rhythm and can be blocked by motor activity or tactile stimulation. The amplitude of the beta rhythm is normally under 30 μV. Similar to the mu rhythm the beta wave may also be enhanced because of a bone defect [33] and also around tumoural regions.
The frequencies above 30 Hz (mainly up to 45 Hz) correspond to the gamma range (sometimes called fast beta wave). Although the amplitudes of these rhythms are very low and their occurrence is rare, detection of these rhythms can be used for confirmation of certain brain diseases. The regions of high EEG frequencies and the highest levels of cerebral blood flow (as well as oxygen and glucose uptake) are located in the frontocentral area. The gamma wave band has also been proved to be a good indication of event-related synchronization (ERS) of the brain and can be used to demonstrate the locus for right and left index finger movement, right toes and the rather broad and bilateral area for tongue movement [36].
Waves of frequencies much higher than the normal activity range of EEG, mostly in the range 200–300 Hz have been found in cerebellar structures of animals but they have not played any role in clinical neurophysiology [37, 38].
Figure 1.6 shows the typical normal brain rhythms with their usual amplitude levels. In general the EEG signals are the projection of neural activities which are attenuated by leptomeninges, cerebrospinal fluid, dura matter, bone, galea, and the scalp. Cortiographic discharges show amplitudes of 0.5–1.5 mV in range and up to several millivolts for spikes. However, on the scalp the amplitudes commonly lie within 10–100 μV.
Figure 1.6 The EEG signal on the top and four typical dominant brain normal rhythms, from low to high frequencies; the delta wave is observed in infants and sleeping adults, the theta wave in children and sleeping adults, the alpha wave is detected in the occipital brain region when there is no attention, and the beta wave appears frontally and parietally with low amplitude
The above rhythms may last if the state of the subject does not change and therefore they are approximately cyclic in nature. On the other hand, there are other brain waveforms, which may:
Often it is difficult to understand and detect the brain rhythms from scalp EEGs, even with trained eyes. Application of advanced signal processing tools, however, should enable separation and analysis of the desired waveforms from within the EEGs. Therefore, definition of foreground and background EEG is very subjective and entirely depends on the abnormalities and applications. We next consider the development in the recording and measurement of EEG signals.
An early model to generate the brain rhythms is the model of Jansen and Rit [42]. This model uses a set of parameters to produce alpha activity through an interaction between inhibitory and excitatory signal generation mechanisms in a single area. The basic idea behind these models is to make excitatory and inhibitory populations interact such that oscillations emerge. This model was later modified and extended to generate and emulate the other main brain rhythms, that is, delta, theta, beta, and gamma, too [43]. The assumptions and mathematics involved in building the Jansen model and its extension are explained in Chapter 3. Application of such models in the generation of postsynaptic potentials and using them as the template to detect, separate, or extract ERPs is of great importance. In Chapter 9 of this book, we can see the use of such a template in the extraction of the ERPs.
1.7 EEG Recording and Measurement
Acquiring signals and images from the human body has become vital for early diagnosis of a variety of diseases. Such data can be in the form of electrobiological signals such as electrocardiogram (ECG) from the heart, electromyography (EMG) from muscles, EEG and MEG from the brain, electrogastrography (EGG) from the stomach, and electro-occlugraphy (electrooptigraphy, EOG) from eye nerves. Measurements can also be taken using ultrasound or radiographs, such as a sonograph (or ultrasound image), computerised tomography (CT), magnetic resonance imaging (MRI) or functional MRI (fMRI), positron emission tomography (PET), single photon emission tomography (SPET), or near-infrared spectroscopy (NIRS).
Functional and physiological changes within the brain may be registered by either EEG, MEG or fMRI. Application of fMRI is, however, very limited in comparison with EEG or MEG due to a number of important reasons:
The first electrical neural activities were registered using simple galvanometers. In order to magnify very fine variations of the pointer a mirror was used to reflect the light projected to the galvanometer on the wall. The d'Arsonval galvanometer later featured a mirror mounted on a movable coil and the light focused on the mirror was reflected when a current passed the coil. The capillary electrometer was introduced by Lippmann and Marey [44]. The string galvanometer, as a very sensitive and more accurate measuring instrument, was introduced by Einthoven in 1903. This became a standard instrument for a few decades and enabled photographic recording.
More recent EEG systems consist of a number of delicate electrodes, a set of differential amplifiers (one for each channel) followed by filters [31], and needle (pen) type registers. The multichannel EEGs could be plotted on plane paper or paper with a grid. Soon after this system came to the market, researchers started looking for a computerised system, which could digitise and store the signals. Therefore, to analyse EEG signals it was soon understood that the signals must be in digital form. This required sampling, quantization, and encoding of the signals. As the number of electrodes grows the data volume, in terms of the number of bits, increases. The computerised systems allow variable settings, stimulations, and sampling frequency, and some are equipped with simple or advanced signal processing tools for processing the signals.
The conversion from analogue to digital EEG is performed by means of multichannel analogue-to-digital converters (ADCs). Fortunately, the effective bandwidth for EEG signals is limited to approximately 100 Hz. For many applications this bandwidth may be considered even half of this value. Therefore, a minimum frequency of 200 Hz (to satisfy the Nyquist criterion) is often enough for sampling the EEG signals. In some applications where a higher resolution is required for representation of brain activities in the frequency domain, sampling frequencies of up to 2000 sample/s may be used.
In order to maintain the diagnostic information the quantization of EEG signals is normally very fine. Representation of each signal sample with up to 16 bits is very popular for the EEG recording systems. This makes the necessary memory volume for archiving the signals massive, especially for sleep EEG and epileptic seizure monitoring records. However, the memory size for archiving the images is often much larger than that used for archiving the EEG signals and with the help of new technology this is a minor problem.
A simple calculation shows that for a one hour recording from 128-electrode EEG signals sampled at 500 samples/s a memory size of 128 × 60 × 60 × 500 × 16 ≈ 3.68 Gbits ≈ 0.45 Gbyte is required. Therefore, for longer recordings of a large number of patients there should be enough storage facilities such as in today's technology Zip disks, CDs, DVDs, large removable hard drives, optical disks, and share servers or the cloud.
Although the format of reading the EEG data may be different for different EEG machines, these formats are easily convertible to spreadsheets readable by most signal processing software packages, such as MATLAB and Python.
The EEG recording electrodes and their proper function are crucial for acquiring high quality data. There are different types of electrodes often used in the EEG recording systems such as:
Disposable (gel-less, and pre-gelled types)Reusable disc electrodes (gold, silver, stainless steel or tin)Headbands and electrode capsSaline-based electrodesNeedle electrodesFor multichannel recordings with a large number of electrodes, electrode caps are often used. Commonly used scalp electrodes consist of Ag-AgCl disks, less than 3 mm in diameter, with long flexible leads that can be plugged into an amplifier. Needle (cortical) electrodes are those which have to be implanted under the skull with minimally invasive operations. High impedance between the cortex and the electrodes as well as electrodes with high impedances can lead to distortion, which can even mask the actual EEG signals. Commercial EEG recording systems are often equipped with impedance monitors. To enable a satisfactory recording the electrode impedances should read less than 5 kΩ and be balanced to within 1 kΩ of each other. For more accurate measurement the impedances are checked after each trial.
Due to the layered and spiral structure of the brain, however, distribution of the potentials over the scalp (or cortex) is not uniform [45]. This may affect some of the results of source localization using the EEG signals.
1.7.1 Conventional EEG Electrode Positioning
The International Federation of Societies for Electroencephalography and Clinical Neurophysiology has recommended the conventional electrode setting (also called 10–20) for 21 electrodes (excluding the earlobe electrodes) as depicted in Figure 1.7 [17]. Often, the earlobe electrodes called A1 and A2, connected respectively to the left and right earlobes, are used as the reference electrodes. The 10–20 system avoids both eyeball placement and considers some constant distances by using specific anatomical landmarks from which the measurement would be made and then uses 10 or 20% of that specified distance as the electrode interval. The odd electrodes are on the left and the even ones on the right.
Figure 1.7 Conventional 10–20 EEG electrode positions for the placement of 21 electrodes
For setting a larger number of electrodes using the above conventional system, the rest of the electrodes are placed equidistantly in between the above electrodes. For example C1 is placed between C3 and Cz. Figure 1.8 represents a larger setting for 75 electrodes, including the reference electrodes based on the guidelines by the American EEG Society. Extra electrodes are sometimes used for the measurement of EOC, ECG, and EMG of the eyelid and eye surrounding muscles. In some applications, such as ERP analysis and brain computer interfacing, a single channel may be used. In such applications, however, the position of the corresponding electrode has to be well determined. For example C3 and C4 can be used to record, respectively, the right and left finger movement related signals for BCI applications. Also F3, F4, P3, and P4 can be used for recordings of the ERP P300 signals.
Figure 1.8 A diagrammatic representation of 10–20 electrode settings for 75 electrodes including the reference electrodes; (a) and (b) represent the three-dimensional measures and (c) indicates a two-dimensional view of the electrode set-up configuration
Two different modes of recordings, namely differential and referential, are used. In the differential mode the two inputs to each differential amplifier are from two electrodes. In referential mode, on the other hand, one or two reference electrodes are used. Several different reference electrode placements can be found in the literature. Physical references can be used as vertex (Cz), linked-ears, linked-mastoids, ipsilateral ear, contralateral ear, C7, bipolar references, and tip of the nose [32]. There are also reference-free recording techniques which actually use a common average reference. The choice of reference may produce topographic distortion if the reference is not relatively neutral. In modern instrumentation, however, the choice of a reference does not play an important role in the measurement [46]. In such systems other references such as FPz, hand, or leg electrodes may be used [47]. The overall setting includes the active electrodes and the references.
In another similar setting called the Maudsley electrode positioning system the conventional 10–20 system has been modified to better capture the signals from epileptic foci in epileptic seizure recordings. The only difference between this system and the 10–20 conventional system is that the outer electrodes are slightly lowered to enable better capturing of the medial temporal seizure signals. The advantage of this system over the conventional one is that it provides a more extensive coverage of the lower part of the cerebral convexity, increasing the sensitivity for the recording from basal sub-temporal structures [48]. Other deviations from the international 10–20 system as used by researchers are found in [49, 50].
In many applications such as brain-computer interfacing (BCI) and study of mental activity, often a small number of electrodes around the movement-related regions are selected and used from the 10–20 setting system.
Figure 1.9 illustrates a typical set of EEG signals during approximately 7 s of normal adult brain activity. It is evident that the dominant signal cycles are those of alpha rhythm.
Figure 1.9 A typical set of EEG signals during approximately 7 s of normal adult brain activity
1.7.2 Conditioning the Signals
Raw EEG signals have amplitudes of the order of μV and contain frequency components of up to 300 Hz. To retain the effective information the signals have to be amplified before the ADC and filtered, either before or after the ADC, to reduce the noise and make the signals suitable for processing and visualization. The filters are designed in such a way as not to introduce any change or distortion to the signals. Highpass filters with cut-off frequency of usually less than 0.5 Hz are used to remove the disturbing very low frequency components such as those of breathing. On the other hand, high frequency noise is mitigated by using lowpass filters with cut-off frequency of approximately 50 to 70 Hz. Use of notch filters with a null frequency of 50 Hz is often necessary to ensure perfect rejection of the strong 50 Hz power supply. In this case the sampling frequency can be as low as twice the bandwidth commonly used by most EEG systems. The commonly used sampling frequencies for EEG recordings are 100, 250, 500, 1000, and 2000 samples/s. The main artefacts can be divided into patient-related (physiological) and system noise. The patient-related or internal artefacts are body movement-related, EMG, ECG (and pulsation), EOG, ballistocardiogram, and sweating. The system artefacts are 50/60 Hz power supply interference, impedance fluctuation, cable defects, electrical noise from the electronic components, and unbalanced impedances of the electrodes. Often in the preprocessing stage these artefacts are highly mitigated and the informative information is restored. Some methods for removing the EEG artefacts will be discussed in the related chapters of this book. Figure 1.10 shows a set of normal EEG signals affected by an eye-blinking artefact. Similarly, Figure 1.11 represents a multichannel EEG set with clear appearance of ECG signals over the electrodes in the occipital region.
Figure 1.10 A set of normal EEG signals affected by eye-blinking artefact
Figure 1.11 A multichannel EEG set with clear appearance of ECG signals over the electrodes in the occipital region
We continue in the next section to highlight the changes in EEG measurements which correlate with physiological and mental abnormalities in the brain.
1.8 Abnormal EEG Patterns
Variations in the EEG patterns for certain states of the subject indicate abnormality. This may be due to distortion and disappearance of abnormal patterns, appearance and increase of abnormal patterns, or disappearance of all patterns. Sharbrough [51] divided the non-specific abnormalities in the EEGs into three categories; (i) widespread intermittent slow wave abnormalities, often in the delta wave range and associated with brain dysfunction, (ii) bilateral persistent EEG usually associated with impaired conscious cerebral reactions, and (iii) focal persistent EEG usually associated with focal cerebral disturbance.
The first category is a burst type signal, which is attenuated by alerting the individual and eye opening, and accentuated with eye closure, hyperventilation, or drowsiness. The peak amplitude in adults is usually localized in the frontal region and influenced by age. In children, however, it appears over the occipital or posterior head region. Early findings showed that this abnormal pattern frequently appears with an increased intracranial pressure with tumour or aqueductal stenosis. Also, it correlates with grey matter disease, both in cortical and subcortical locations. However, it can be seen in association with a wide variety of pathological processes, varying from systemic toxic or metabolic disturbances to focal intracranial lesions.
Regarding the second category, that is, bilateral persistent EEG, the phenomenon in different stages of impaired, conscious, purposeful responsiveness are aetiologically non-specific and the mechanisms responsible for their generation are only partially understood. However, the findings in connection with other information concerning aetiology and chronicity may be helpful in arriving more quickly at an accurate prognosis concerning the patient's chance of recovering his previous conscious life.
