Understanding the human brain has been one of the most challenging topics in neuroscience and science in general. Different and heterogeneous paths to extracting valuable information and knowledge have been proposed over the past years in aggregate trying to solve one of the biggest mysteries: How does the brain work? But is our brain capable of understanding itself? Fortunately, in this book, we do not attempt to complete such an obscure quest, but we can certainly contribute in this endeavor by reviewing modern quantitative analysis methodologies.A\Data acquisition and analysis techniques have developed in parallel, enabling the measurement of connectivity between brain regions. Anatomic and functional neuroimag-ing methods for detecting brain activity using both electroencephalography (EEG) and magnetoencephalography (MEG) have developed dramatically during the past two decades. Although neuroimaging studies based on positron emission tomography (PET) and functional magnetic resonance imaging (fMRI) are especially popular mostly due to the prospective possibility of being able to measure the neuronal activity on a high temporal as well as spatial scale, there is no physical way (several hundreds of milliseconds needed to reflect blood oxygen level dependent—BOLD—effects) of reaching the ultrahigh temporal precision of the EEG and MEG techniques that are already widely used. In this sense, EEG and MEG remain of paramount importance in the neuroimaging community and may be used as single methods or in combination with simultaneous fMRI scans as discus ed in the final two chapters of this book.A\A number of connectivity analysis methodologies addressing a wide variety of clinical applications including epilepsy, schizophrenia, Alzheimer's disease, and alcoholism, as well as cognitive studies are presented in this book, each one having its specific strengths and weaknesses. Our intention is to provide a comprehensive overview of the most modern and widely established approaches mainly applied in, but not limited to, decomposing high-resolution multichannel EEG and MEG signals into functional interconnected brain regions, as the functional segregation and integration concept suggests. We wish to aid interested researchers by including manuscripts describing the theoretical basis of each presented method along with prosperous application domains, in the form of a balanced mixture of theoretical tutorials, comprehensive reviews, and original research. Emphasis is given on the underlying assumptions, on technical matters that greatly affect the outcome of each proposed method, on the ambitions, and on the domain of application of each method. Furthermore, links to graph theory and visualization of connectivity motifs are also presented in an attempt to better describe the functional characteristics of brain networks.A\Chapter Operational Architectonics Methodology for EEG Analysis: Theory and Results presents the strengths and drawbacks of the EEG signals. Only a deep understanding of brain spatiotemporal dynamics guaranties genuine long-term progress in psycho-physiological, cognitive, and medical science. Andrew Fingelkurts and Alexander Fingelkurts introduce various aspects of operational architectonics, a methodology addressing the peculiarities of spatial and temporal EEG nonstationarity. Such a novel technique is sensitive to the underlying quasi-stationary nature of EEG signal and plausible for a better understanding of the functional organization of the neocortex and its relation to consciousness. Thus, there is a wide application domain, including epilepsy, schizophrenia, major depression, sleep disorders, and chronic opioid abuse, in the clinical field reflecting mostly neuropsychiatric disorders. Design and administration of psychotropic drugs are also discussed, such as the methadone effects as a maintenance treatment for heroin-dependent patients and Lorazepam administration. Cognitive neuroscience applications include Working Memory experiments and ontological and personality development studies, while "operational architectonics" touches also upon neurophilosophy and artificial intelligence.A\Chapters Clinical Electroencephalography in the Diagnosis and Management of Epilepsy, Effective Brain Connectivity from Intracranial EEG Recordings: Identification of Epileptogenic Zone in Human Focal Epilepsies, On the Effect of Volume Conduction on Graph Theoretic Measures of Brain Networks in Epilepsy, and Methods for Seizure Detection and Prediction: An Overview focus on different methodologies applied in epilepsy. More specifically, Chapter Clinical Electroencephalography in the Diagnosis and Management of Epilepsy provides an introductory clinical perspective of EEG in the diagnosis and management of epilepsy, overviewing the present status and emerging approaches.A\Chapter Effective Brain Connectivity from Intracranial EEG Recordings: Identification of Epileptogenic Zone in Human Focal Epilepsies provides an overview of the different intracranial EEG signal processing methods used to identify the epileptogenic zone that may be resected surgically to suppress seizures. Particular attention is being given to the methods aimed at characterizing effective brain connectivity using intracranial EEG recordings. G. Varotto and colleagues present connectivity pattern analysis associated with a particular form of focal epilepsy (type II focal cortical dysplasia), based on multivariate autoregressive parametric models and measures derived from graph theory.A\Chapter On the Effect of Volume Conduction on Graph Theoretic Measures of Brain Networks in Epilepsy evaluates and compares two standard and most commonly used linear connectivity measures—cross-correlation in the time domain and coherence in the frequency domain—with measures that account for volume conduction, namely corrected cross-correlation, imaginary coherence, phase lag index, and weighted phase lag index. M. Christodoulakis et al. focus mosdy on the way connectivity measures are affected by both volume conduction and the choice of recording reference (montage) in the time and frequency domain. Graph-theoretic indices are again used to assess network topology changes in epileptic subjects.A\Chapter Methods for Seizure Detection and Prediction: An Overview discusses another application of linear/nonlinear analysis, chaos, and information-based analysis, in detecting and even predicting epilepsy. G. Giannakakis et al. review feature selection procedures from the aforementioned methods that are able to detect and classify epileptic states. Each method's accuracy is evaluated through performance measures indicating the strengths of each proposed technique.A\Functional connectivity measures and, more specifically, coherence, phase synchronization, and nonlinear state-space generalized synchronization assessment methods are further discussed in Chapter Graph-theoretic Indices of Evaluating Brain Network Synchronization: Application in an Alcoholism Paradigm. Synchronization matrices define graphs whose topological structure and properties arc characterized using measures for graphs and weighted networks. Graph-theoretic measures are also used as the tools to visualize and characterize the topology of a brain network as in a working memory task-related alcoholism paradigm.A\Chapter Time-Varying Effective Connectivity for Investigating the Neurophysiological Basis of Cognitive Processes describes the methodological advancements developed during the last 20 years in the field, of effective connectivity based on Granger causality and linear autoregressive modeling. Apart from introducing the most widely accepted methodologies, a graph-theoretic approach demonstrates their potential application in a motor imagery process.A\Chapter Assessment of Sensory Gating Deficit in Schizophrenia Using a Wavelet Transform Methodology on Auditory Paired-Click Evoked Potentials discusses the use of EEG in the search for biomarkers of schizophrenia and, more specifically, the assessment of the sensory gating process in patients with schizophrenia via time-frequency decomposition of the EEG signals.A\G. Zouridakis and colleagues in Chapter Schizophrenia Assessment Using Single-Trial Analysis of Brain Activity discuss the clinical application of Independent Component Analysis (ICA) with a focus on schizophrenia. The methodology presented is based on single-trial analysis using an iterative independent component analysis procedure. This method is capable of identifying and measuring the amplitude, latency, and overall morphology of individual components in single trials, and as such, permits the study of phase characteristics among single trials, while preserving known features of the average evoked potentials.A\Phase synchronization and two recent adaptations of the Common Spatial Patterns method are further investigated in Chapter Phase Variants of the Common Spatial Patterns Method, provided by T. Camilleri et al. There is no one best methodology for estimating phase synchronization. The selection of the techniques presented complements each other and is evaluated in a motor imagery task.A\Chapter Estimation of Regional Activation Maps and Interdependencies from Minimum Norm Estimates of Magnetoencephalography (MEG) Data describes the development of a minimally supervised pipeline for the analysis of event-related MEG recordings that preserves the temporal resolution of the data and enables estimation of patterns of regional interdependencies between activated regions in an object naming task. The proposed approach is based on a spatiotemporal source-clustering algorithm initially applied to identify extended regions of significant activation, and subsequently regional interdependencies are estimated through cross lag correlation analysis between time series representing the time course of activity within each cluster.A\Chapter Blind Signal Separation Methods in Computational Neuroscience presents a survey of Blind Source Separation methods based on Independent and Sparse Component Analysis. The theoretical basis and mathematical formulation of the most widely adopted methods are described. M. N. Syed et al. evaluate the strengths and weaknesses of different formulations in an experimental setup capturing simultaneous electrical activity over the scalp (EEG) and over the exposed surface of the cortex (ECoG). Since the data from this experiment is simultaneously collected from above the scalp and under the scalp, it opens the door to understand the mixing mechanism across the brain. Further applications in fMRI, MRI, and finger prints are also mentioned.A\Chapter Current Trends in ERP Analysis Using EEG and EEG/fMRI Synergistic Methods introduces methods and measures used for the analysis of Event Potentials but combines the excellent temporal resolution of the EEG with the spatial details provided by the fMRI. Such an approach allows moving beyond isolation and connection of specific EEG features to specific cognitive processes. Fusion between EEG and fMRI looks very promising since the strong feature of the one is the weakness of the other. Reality though proved more complex, and EEG and fMRI fusion is still an open area for research.A\Finally, Chapter Computer-Based Assessment of Alzheimer's Disease Employing fMRI and/or EEG: A Comprehensive Review focuses on assessing Alzheimer's disease using both fMRI and EEG approaches. In addition, fusion of both approaches is also presented.A\Over the years, EEG and more recently MEG have evolved and are widely accepted in today's clinical practice. However, only traditional analysis techniques are mostly used. We believe that in the coming years most of the emerging and promising techniques presented in this book could become more established and reach the clinical research community.A\Hopefully this book, touching upon both the biomedical and computational aspects of this exciting and rapidly evolving field, will be an interesting and enjoyable read and will allow for a more in depth understanding of the brain's underlying mechanisms. Heraklion, Crete, Greece: Vangelis Sakkalis

Preface to the Series v

Preface vii

Acknowledgaments xi

About the Editor xiii

Contributors xvii

Operational Architectonics Methodology for EEG Analysis: Theory and Results 1

Clinical Electroencephalography in the Diagnosis and Management of Epilepsy 61

Effective Brain Connectivity from Intracranial EEG Recordings: Identification of Eplleptogenic Zone in Human Focal Epilepsies 87

On the Effect of Volume Conduction on Graph Theoretic Measures of Brain Networks In Epilepsy 103

Methods for Seizure Detection and Prediction: An Overview 131

Graph-theoretic Indices of Evaluating Brain Network Synchronization: Application in an Alcoholism Paradigm 159

Time-Varying Effective Connectivity for Investigating the Neurophysiological Basis of Cognitive Processes 171

Assessment of Sensory Gating Deficit in Schizophrenia Using a Wavelet Transform Methodology on Auditory Paired-Click Evoked Potentials 205

Schizophrenia Assessment Using Single-THal Analysis of Brain Activity 231

Phase Variants of the Common Spatial Patterns Method 249

Estimation of Regional Activation Maps and Interdependencies from Minimum Norm Estimates of Magnetoencephalography (MEG) Data 267

Blind Signal Separation Methods in Computational Neuroscience 291

Current It-ends in ERP Analysis Using EEG and EEG/fMRI Synergistic Methods 323

Computer-Based Assessment of Alzheimer's Disease Employing fMRI and/or EEG: A Comprehensive Review 351

Index 285

中国医科院医学信息研究所