Atjaunināt sīkdatņu piekrišanu

Handbook of Neuroimaging Data Analysis [Hardback]

Edited by , Edited by (University of California, Irvine, USA), Edited by (Primary affiliation: Institute of Biological Psychiatry, Denmark; Secondary affiliation: University of California, San Diego, La Jolla, USA-), Edited by (Johns Hopkins University, Baltimore, Maryland, USA)
  • Formāts: Hardback, 702 pages, height x width: 254x178 mm, weight: 1680 g, 9 Tables, black and white; 172 Illustrations, black and white
  • Sērija : Chapman & Hall/CRC Handbooks of Modern Statistical Methods
  • Izdošanas datums: 14-Nov-2016
  • Izdevniecība: Chapman & Hall/CRC
  • ISBN-10: 1482220970
  • ISBN-13: 9781482220971
Citas grāmatas par šo tēmu:
  • Hardback
  • Cena: 249,78 €
  • Grāmatu piegādes laiks ir 3-4 nedēļas, ja grāmata ir uz vietas izdevniecības noliktavā. Ja izdevējam nepieciešams publicēt jaunu tirāžu, grāmatas piegāde var aizkavēties.
  • Daudzums:
  • Ielikt grozā
  • Piegādes laiks - 4-6 nedēļas
  • Pievienot vēlmju sarakstam
Handbook of Neuroimaging Data AnalysisPalielināt
  • Formāts: Hardback, 702 pages, height x width: 254x178 mm, weight: 1680 g, 9 Tables, black and white; 172 Illustrations, black and white
  • Sērija : Chapman & Hall/CRC Handbooks of Modern Statistical Methods
  • Izdošanas datums: 14-Nov-2016
  • Izdevniecība: Chapman & Hall/CRC
  • ISBN-10: 1482220970
  • ISBN-13: 9781482220971
Citas grāmatas par šo tēmu:
Permanent link: https://www.kriso.lv/db/9781482220971.html
Keywords:

This book explores various state-of-the-art aspects behind the statistical analysis of neuroimaging data. It examines the development of novel statistical approaches to model brain data. Designed for researchers in statistics, biostatistics, computer science, cognitive science, computer engineering, biomedical engineering, applied mathematics, physics, and radiology, the book can also be used as a textbook for graduate-level courses in statistics and biostatistics or as a self-study reference for Ph.D. students in statistics, biostatistics, psychology, neuroscience, and computer science.

Recenzijas

"Handbook of Neuroimaging Data Analysis is a great source to help you get started . . . If you find a particular modality that interests you, just email one of the authors in the book who also works on data analysis within that modality. They are all friendly and helpful, and they will point you to sources of publically available data." ~Timothy D. Johnson

"These chapters are primarily written by statisticians, but the book is nicely balanced by contributions from biomedical engineers, psychologists, and cognitive scientists. . . I recommend this book to statisticians interested in learning about neuroimaging and contributing to its growth." ~Journal of the American Statistical Association

List of Figures xix
List of Tables xxxiii
Preface xxxv
Contributors xxxvii
I Overview 1(14)
1 Overview of the Handbook
3(12)
1.1 Introduction
3(1)
1.2 A Brief History of Neuroimaging
4(1)
1.3 Modalities
4(3)
1.4 Statistical Methods
7(4)
1.4.1 Preprocessing
8(1)
1.4.2 Methods in Structural Neuroimaging
8(1)
1.4.3 Localizing Areas of Activation
9(1)
1.4.4 Brain Connectivity
9(1)
1.4.5 Analysis of Electroencephalograms
10(1)
1.4.6 Multi-Modal Analysis
10(1)
1.5 Additional Topics
11(1)
1.5.1 Meta-Analysis
11(1)
1.5.2 Experimental Design in fMRI
11(1)
1.5.3 Experimental Design in EEG
11(1)
1.5.4 Imaging Genetics
12(1)
1.6 Conclusions
12(1)
Bibliography
13(2)
II Imaging Modalities 15(188)
2 Positron Emission Tomography: Some Analysis Methods
17(18)
John Aston
2.1 Introduction
17(2)
2.2 Background
19(1)
2.3 Tracer Kinetic Modelling: Compartmental Approaches
20(4)
2.3.1 Plasma Input Functions Models
20(3)
2.3.2 Reference Tissue Models
23(1)
2.4 Estimation and Statistical Methods
24(3)
2.4.1 Non-Linear Least Squares
24(1)
2.4.2 Basis Function Methods
25(1)
2.4.3 Model Selection
26(1)
2.5 Other Modelling Approaches
27(2)
2.5.1 Graphical Methods
27(1)
2.5.2 Bayesian Approaches
28(1)
2.5.3 Non-Parametric Approaches
28(1)
2.6 Further Modelling Considerations
29(1)
Bibliography
30(5)
3 Structural Magnetic Resonance Imaging
35(30)
Wesley K. Thompson
Hauke Bartsch
Martin A. Lindquist
3.1 Introduction
36(1)
3.2 Image Acquisition
37(3)
3.2.1 MRI Physics
38(1)
3.2.2 Image Reconstruction
38(1)
3.2.3 MRI Sequences
38(1)
3.2.4 MRI Artifacts
39(1)
3.3 Multi-Subject MRI
40(11)
3.3.1 Registration
40(1)
3.3.1.1 Volumetric Registration
41(1)
3.3.1.2 Surface-Based Registration
41(2)
3.3.2 Segmentation
43(1)
3.3.2.1 Foreground from Background Segmentation
43(1)
3.3.2.2 Brain Tissue Segmentation
44(1)
3.3.3 Templates and Atlases
45(1)
3.3.4 Morphometry
46(1)
3.3.4.1 Subcortical Volumes
46(1)
3.3.4.2 Voxel-Based Morphometry
47(1)
3.3.4.3 Deformation- and Tensor-Based Morphometry
47(1)
3.3.4.4 Surface-Based Measures
48(1)
3.3.4.5 Other Morphometric Measures
49(1)
3.3.5 Statistical Analyses
50(1)
3.3.5.1 Statistical Parametric Maps
51(1)
3.4 Miscellaneous Topics
51(2)
3.4.1 Structural Integrity and Tumor Detection
51(1)
3.4.2 Anatomical References for Functional Imaging
52(1)
3.4.3 Multi-Center Studies
52(1)
3.4.4 Imaging Genetics
53(1)
3.5 Glossary of MRI Terms
53(1)
Bibliography
54(11)
4 Diffusion Magnetic Resonance Imaging (dMRI)
65(44)
Jian Cheng
Hongtu Zhu
4.1 Introduction to Diffusion MRI
66(4)
4.1.1 Diffusion Weighted Imaging (DWI)
66(1)
4.1.1.1 Diffusion Gradient Sequence
66(1)
4.1.1.2 Free Diffusion
67(1)
4.1.1.3 Restricted Diffusion
67(2)
4.1.2 Diffusion Tensor Imaging (DTI)
69(1)
4.1.2.1 Scalar Indices and Eigenvectors of Diffusion Tensor
69(1)
4.2 High Angular Resolution Diffusion Imaging (HARDI)
70(9)
4.2.1 Generalization of Diffusion Tensor Imaging
70(1)
4.2.1.1 Mixture of Tensor Model
70(1)
4.2.1.2 Generalized DTI (GDTI)
71(1)
4.2.1.3 High-Order Tensor Model, ADC-Based Model
72(1)
4.2.2 Diffusion Spectrum Imaging (DSI)
73(1)
4.2.3 Hybrid Diffusion Imaging (HYDI)
74(1)
4.2.4 Q-Ball Imaging (QBI)
75(1)
4.2.4.1 Original Q-Ball Imaging
75(1)
4.2.4.2 Exact Q-Ball Imaging
76(1)
4.2.5 Diffusion Orientation Transform (DOT)
77(1)
4.2.6 Spherical Deconvolution (SD)
77(1)
4.2.7 Diffusion Propagator Imaging (DPI)
77(1)
4.2.8 Simple Harmonic Oscillator Reconstruction and Estimation (SHORE)
78(1)
4.2.9 Spherical Polar Fourier Imaging (SPFI)
78(1)
4.3 Reconstruction
79(4)
4.3.1 Noise Components and Voxelwise Estimation Methods
79(1)
4.3.2 Spatial-Adaptive Estimation Methods
80(3)
4.4 Tractography Algorithms
83(3)
4.5 Uncertainty in Estimated Diffusion Quantities
86(1)
4.6 Sampling Mechanisms
87(2)
4.7 Registration
89(1)
4.8 Group Analysis
90(3)
4.9 Public Resources
93(2)
4.9.1 D at asets
93(1)
4.9.2 Software
94(1)
4.10 Glossary
95(1)
Bibliography
96(13)
5 A Tutorial for Multisequence Clinical Structural Brain MRI
109(30)
Ciprian Crainiceanu
Elizabeth M. Sweeney
Ani Eloyan
Russell T. Shinohara
5.1 Introduction
110(4)
5.1.1 What Are These Images?
112(1)
5.1.2 How Can We Handle sMRI?
113(1)
5.1.3 What Are Some Major Pitfalls When Starting Working on sMRI?
114(1)
5.2 Data Structure and Intuitive Description of Associated Problems
114(1)
5.3 Acquisition and Reconstruction
115(1)
5.4 Preprocessing
116(13)
5.4.1 Inhomogeneity Correction
117(1)
5.4.1.1 Concepts
118(1)
5.4.1.2 Practical Approaches, Software, and Application to Data
118(1)
5.4.2 Skull Stripping
119(1)
5.4.2.1 Concepts
119(1)
5.4.2.2 Practical Approaches, Software, and Application to Data
119(1)
5.4.3 Interpolation
119(1)
5.4.3.1 Concepts
120(1)
5.4.3.2 Practical Approaches, Software, and Application to Data
121(1)
5.4.4 Spatial Registration
121(1)
5.4.4.1 Concepts
121(1)
5.4.4.2 Practical Approaches, Software, and Application to Data
123(3)
5.4.5 Intensity Normalization
126(1)
5.4.5.1 Concepts
127(1)
5.4.5.2 Practical Approaches, Software, and Application to Data
127(2)
5.5 Analysis
129(4)
5.5.1 Lesion Segmentation
129(1)
5.5.2 Lesion Mapping
130(1)
5.5.3 Longitudinal and Cross Sectional Intensity Analysis
131(2)
5.6 Conclusions
133(1)
Bibliography
133(6)
6 Principles of Functional Magnetic Resonance Imaging
139(36)
Martin A. Lindquist
Tor D. Wager
6.1 Introduction
139(2)
6.2 The Basics of fMRI Data
141(7)
6.2.1 Principles of Magnetic Resonance Signal Generation
141(1)
6.2.1.1 The MRI Scanner
141(1)
6.2.1.2 Basic MR Physics
142(1)
6.2.1.3 Image Contrast
143(1)
6.2.2 Image Formation
144(2)
6.2.3 From MRI to fMRI
146(2)
6.3 BOLD fMRI
148(5)
6.3.1 Understanding BOLD fMRI
148(2)
6.3.2 Spatial Limitations
150(1)
6.3.3 Temporal Limitations
151(1)
6.3.4 Acquisition Artifacts
152(1)
6.4 Modeling Signal and Noise in fMRI
153(4)
6.4.1 BOLD Signal
153(2)
6.4.2 Noise and Nuisance Signal
155(2)
6.5 Experimental Design
157(1)
6.6 Preprocessing
158(2)
6.7 Data Analysis
160(5)
6.7.1 Localization
160(2)
6.7.2 Connectivity
162(2)
6.7.3 Prediction
164(1)
6.8 Resting-State fMRI
165(1)
6.9 Data Format, Databases, and Software
166(2)
6.10 Future Developments
168(1)
Bibliography
169(6)
7 Electroencephalography (EEG): Neurophysics, Experimental Methods, and Signal Processing
175(28)
Michael D. Nunez
Paul L. Nunez
Ramesh Srinivasan
7.1 Introduction
175(2)
7.2 The Neurophysics of EEG
177(3)
7.3 Synchronization and EEG
180(2)
7.4 Recording EEG
182(1)
7.5 Preprocessing EEG
183(1)
7.6 Artifact Removal
184(3)
7.7 Stationary Data Analysis
187(7)
7.8 Nonstationary Data Analysis
194(3)
7.9 Summary
197(1)
Bibliography
197(6)
III Statistical Methods And Models 203(424)
8 Image Reconstruction in Functional MRI
205(28)
Daniel B. Rowe
8.1 Introduction
205(1)
8.2 The Fourier Transform
206(8)
8.2.1 One-Dimensional Fourier Transform
206(4)
8.2.2 Two-Dimensional Fourier Transform
210(4)
8.3 FMRI Acquisition and Reconstruction
214(6)
8.3.1 The Signal Equation and k-Space Coverage
214(3)
8.3.2 Nyquist Ghost k-Space Correction
217(3)
8.4 Image Processing
220(8)
8.4.1 Reconstruction Isomorphism Representation
220(2)
8.4.2 Image Processing Implications
222(6)
8.5 Additional Topics and Discussion
228(2)
8.5.1 Complex-Valued fMRI Activation
229(1)
8.5.2 Discussion
230(1)
Acknowledgments
230(1)
Bibliography
230(3)
9 Statistical Analysis on Brain Surfaces
233(30)
Moo K. Chung
Seth D. Pollak
Jamie L. Hanson
9.1 Introduction
234(1)
9.2 Surface Parameterization
235(4)
9.2.1 Local Parameterization by Quadratic Polynomial
236(1)
9.2.2 Surface Flattening
236(1)
9.2.3 Spherical Harmonic Representation
237(2)
9.3 Surface Registration
239(2)
9.3.1 Affine Registration
239(1)
9.3.2 SPHARM Correspondence
239(2)
9.3.3 Diffeomorphic Registration
241(1)
9.4 Cortical Surface Features
241(2)
9.4.1 Cortical Thickness
241(1)
9.4.2 Surface Area and Curvatures
242(1)
9.4.3 Gray Matter Volume
242(1)
9.5 Surface Data Smoothing
243(5)
9.5.1 Diffusion Smoothing
243(2)
9.5.2 Iterated Kernel Smoothing
245(1)
9.5.3 Heat Kernel Smoothing
246(2)
9.6 Statistical Inference on Surfaces
248(6)
9.6.1 General Linear Models
248(1)
9.6.2 Multivariate General Linear Models
249(1)
9.6.3 Small-n Large-p Problems
250(1)
9.6.4 Longitudinal Models
251(1)
9.6.5 Random Field Theory
252(2)
Bibliography
254(9)
10 Neuroimage Preprocessing
263(46)
Stephen C. Strother
Nathan Churchill
10.1 Introduction
264(3)
10.2 Principles for Studying and Optimizing Preprocessing Pipelines
267(3)
10.2.1 The Utility of Simulated Datasets
268(1)
10.2.2 Quantifying the Impact of Preprocessing Changes
268(1)
10.2.3 The Neuroscientific Importance of Preprocessing Choices
269(1)
10.3 Metrics for Evaluating Neuroimaging Pipelines
270(7)
10.3.1 Pseudo-ROC Curves
271(1)
10.3.2 Cluster Overlap Metrics
271(1)
10.3.3 Intra-Class Correlation Coefficient
272(3)
10.3.4 Spatial Pattern Reproducibility Using Correlations
275(1)
10.3.5 Similarity Metric Ranking Approaches
275(1)
10.3.6 Prediction Metrics
276(1)
10.3.7 Combined Prediction versus Spatial Reproducibility Metrics
276(1)
10.4 Preprocessing Pipeline Testing in the Literature
277(3)
10.4.1 Between-Subject, MRI Brain Registration
277(1)
10.4.2 Preprocessing for fMRI Resting-State Analysis
278(2)
10.4.3 Preprocessing for fMRI Task-Based Analysis
280(1)
10.5 A Case Study: Optimizing fMRI Task Preprocessing
280(13)
10.5.1 Optimization with Prediction and Reproducibility Metrics
280(4)
10.5.2 An Individual Subject's (P, R) Curves
284(1)
10.5.3 fMRI Preprocessing Pipeline Optimization with (P, R) Curves
285(1)
10.5.3.1 Some fMRI Datasets
285(1)
10.5.3.2 Selecting Optimal Preprocessing Pipeline Steps
286(3)
10.5.4 Fixed and Individually Optimized Preprocessing Pipelines
289(3)
10.5.5 Independent Tests of Pipeline Optimization Results
292(1)
10.6 Discussion of Open Problems and Pitfalls
293(2)
Bibliography
295(14)
11 Linear and Nonlinear Models for fMRI Time Series Analysis
309(26)
Tingting Zhang
Haipeng Shen
Fan Li
11.1 Introduction
309(1)
11.2 The GLM: Single-Level Analysis
310(2)
11.3 Modeling the Hemodynamic Response Function in the Time Domain
312(9)
11.3.1 Parametric Models
313(4)
11.3.2 Nonparametric Models
317(3)
11.3.3 Comparison of Different HRF Estimation Methods
320(1)
11.4 Hemodynamic Response Estimation in the Frequency Domain
321(1)
11.5 Multi-Subject Analysis
322(1)
11.5.1 Semi-Parametric Approaches
323(1)
11.6 Nonlinear Models
323(4)
11.6.1 The Balloon Model
324(1)
11.6.2 Volterra Series Model
325(1)
11.6.3 Bi-Exponential Nonlinear Model
326(1)
11.6.4 Volterra Series Models for Multi-Subject Data
326(1)
11.7 Summary and Future Directions
327(1)
Bibliography
328(7)
12 Functional Neuroimaging Group Studies
335(20)
Bertrand Thirion
12.1 Introduction
335(2)
12.2 Variability of Brain Shape and Function
337(2)
12.3 Mixed-Effects and Fixed-Effects Analyses
339(1)
12.4 Group Analysis for Functional Neuroimaging
339(5)
12.4.1 Problem Setting and Notations
340(1)
12.4.2 Estimation
341(1)
12.4.3 Statistical Inference
342(1)
12.4.4 The Random Effects t-Test
343(1)
12.5 Taking into Account the Spatial Context in Statistical Inference
344(2)
12.6 Type I Error Control with Permutation Testing
346(1)
12.7 Illustration of Various Inference Strategies on an Example
347(3)
12.8 Conclusion
350(1)
Bibliography
350(5)
13 Corrections for Multiplicity in Functional Neuroimaging Data
355(14)
Nicole A. Lazar
13.1 Introduction
355(1)
13.2 Control of Familywise Error Rate
356(4)
13.3 Control of False Discovery Rate
360(2)
13.4 Accounting for Spatial Dependence
362(2)
13.5 Summary
364(1)
Bibliography
365(4)
14 Functional Connectivity Analyses for fMRI Data
369(30)
Ivor Cribben
Mark Fiecas
14.1 Introduction
369(2)
14.2 Methods and Measures for FC
371(5)
14.2.1 Setup
371(1)
14.2.2 Cross-Correlation and Partial Cross-Correlation
371(1)
14.2.3 Stability Selection
372(1)
14.2.4 Cross-Coherence and Partial Cross-Coherence
373(1)
14.2.5 Mutual Information
374(1)
14.2.6 Principal and Independent Components Analyses
374(1)
14.2.7 Time-Varying Connectivity
375(1)
14.3 Simulation Study
376(5)
14.3.1 Results
378(3)
14.4 Functional Connectivity Analysis of Resting-State fMRI Data
381(3)
14.4.1 Data Description and Preprocessing
381(1)
14.4.2 Overview of the Estimation Procedure
382(1)
14.4.3 Results
382(2)
14.5 Future Directions and Open Problems
384(5)
Bibliography
389(10)
15 Multivariate Decompositions in Brain Imaging
399(20)
Ani Eloyan
Vadim Zipunnikov
Juemin Yang
Brian Caffo
15.1 Introduction
399(1)
15.2 Principal Component Analysis and Singular Value Decomposition
400(3)
15.2.1 Singular Value Decomposition
401(1)
15.2.2 Principal Components Analysis
401(1)
15.2.3 PCA in Brain Imaging
402(1)
15.3 Structured PCA Models
403(2)
15.3.1 Calculation of High-Dimensional PCA
404(1)
15.4 Independent Component Analysis
405(4)
15.4.1 ICA in Brain Imaging
406(1)
15.4.2 Homotopic Group ICA
407(1)
15.4.3 Computation of High-Dimensional ICA
407(2)
15.5 Discussion of Other Methods
409(2)
15.6 Acknowledgements
411(1)
Bibliography
411(8)
16 Effective Connectivity and Causal Inference in Neuroimaging
419(22)
Martin A. Lindquist
Michael E. Sobel
16.1 Introduction
419(1)
16.2 Effective Connectivity
420(2)
16.3 Models of Effective Connectivity
422(10)
16.3.1 Structural Equation Models
422(5)
16.3.2 Dynamic Causal Models
427(2)
16.3.3 Granger Causality
429(3)
16.4 Effective Connectivity and Causation
432(3)
16.5 Conclusions
435(1)
Bibliography
436(5)
17 Network Analysis
441(26)
Cedric E. Ginestet
Mark Kramer
Eric D. Kolaczyk
17.1 Introduction
441(2)
17.2 Network Construction
443(4)
17.2.1 Notation
443(1)
17.2.2 Vertex Set
444(2)
17.2.3 Edge Set
446(1)
17.2.4 Thresholding Networks
446(1)
17.3 Descriptive Measures of Network Topology
447(3)
17.3.1 Characteristic Path Length
447(1)
17.3.2 Clustering Coefficient
448(1)
17.3.3 Degree Distribution
449(1)
17.4 Network Models
450(5)
17.4.1 Erdos-Renyi Random Graphs
451(1)
17.4.2 Small-World Networks
452(1)
17.4.3 Preferential Attachment
453(1)
17.4.4 Exponential Random Graph Models
453(1)
17.4.5 Stochastic Block Models
454(1)
17.5 Estimation and Comparison of Networks
455(5)
17.5.1 Statistical Parametric Networks
456(1)
17.5.2 Density-Integrated Topology
457(1)
17.5.3 Comparison of Weighted Networks
458(2)
17.6 Conclusion
460(1)
Bibliography
460(7)
18 Modeling Change in the Brain: Methods for Cross-Sectional and Longitudinal Data
467(28)
Philip T. Reiss
Ciprian M. Crainiceanu
Wesley K. Thompson
Lan Huo
18.1 Introduction
468(1)
18.2 Notation and Road Map
468(1)
18.3 Cross-Sectional and Longitudinal Designs
469(3)
18.3.1 Cross-Sectional Designs
470(1)
18.3.2 Single-Cohort Longitudinal Designs
470(1)
18.3.3 Multi-Cohort Longitudinal Designs
471(1)
18.4 Region-Wise Linear Models for the Mean
472(5)
18.4.1 What Is v, and What Is t?
472(1)
18.4.2 Cross-Sectional Data
473(1)
18.4.3 Longitudinal Data
473(1)
18.4.3.1 Mixed-Effects Models
473(1)
18.4.3.2 Marginal Models
474(1)
18.4.4 Relative Efficiency for Estimating the Mean Function
474(1)
18.4.5 Complications Due to Misalignment
475(2)
18.4.6 Borrowing Information "Spatially"
477(1)
18.5 Nonlinear Models for the Mean
477(5)
18.5.1 Polynomial Models
477(1)
18.5.2 Nonparametric and Semiparametric Models
477(3)
18.5.3 Analyses with Repeated Cross-Sectional Subsamples
480(2)
18.6 Beyond Modeling the Mean
482(4)
18.6.1 Individual-Specific Curves
482(1)
18.6.2 Modeling Components of Change: Longitudinal Functional Principal Component Analysis
483(2)
18.6.3 Modeling the Entire Age-Specific Distribution
485(1)
18.6.4 Modeling Local Rates of Change
485(1)
18.7 Discussion
486(1)
Bibliography
487(8)
19 Joint fMRI and DTI Models for Brain Connectivity
495(28)
F. DuBois Bowman
Sean Simpson
Daniel Drake
19.1 Brain Connectivity
495(2)
19.1.1 Structural Connectivity
496(1)
19.1.2 Functional Connectivity
496(1)
19.2 Single Modality Methods
497(8)
19.2.1 Methods for Functional Connectivity
497(1)
19.2.1.1 Defining the Spatial Scale for Connectivity Analysis
497(1)
19.2.1.2 Measures of Association
498(1)
19.2.1.3 Modeling Approaches
499(1)
19.2.1.4 Partitioning Methods
500(1)
19.2.1.5 Network Methods
500(2)
19.2.2 Methods for Effective Connectivity
502(1)
19.2.3 Determining Structural Connectivity
502(1)
19.2.3.1 Diffusion Weighted Imaging and DTI
503(1)
19.2.3.2 Tractography
504(1)
19.3 Multimodal Approaches
505(8)
19.3.1 Sequential Procedures
506(2)
19.3.2 Functional Connectivity with Anatomical Weighting
508(1)
19.3.3 Modeling Joint Activation and Structural Connectivity
509(1)
19.3.3.1 Functional Coherence
510(1)
19.3.3.2 Ascendancy
511(1)
19.3.3.3 Likelihood Function
511(1)
19.3.4 Joint ICA
512(1)
19.3.5 Multimodal Prediction Methods
512(1)
19.4 Conclusion
513(2)
Bibliography
515(8)
20 Statistical Analysis of Electroencephalograms
523(44)
Yuxiao Wang
Lechuan Hu
Hernando Ombao
20.1 Introduction
524(1)
20.2 Spectral Analysis of a Single-Channel EEG
524(17)
20.2.1 Brief Description of the Data
525(1)
20.2.2 Fourier-Domain Approach
525(1)
20.2.2.1 The Fourier Regression Model and Variance Decomposition
525(1)
20.2.2.2 The Spectrum of a Single-Channel Time Series
528(1)
20.2.2.3 Estimating the Spectrum via Periodograms
528(1)
20.2.2.4 Other Periodogram-Based Estimation Methods
531(1)
20.2.2.5 Examples of Smoothing Periodograms
531(1)
20.2.2.6 Multitaper Method (MTM)
531(1)
20.2.3 Time-Domain Approach
532(1)
20.2.3.1 Moving Average (MA) Model
532(1)
20.2.3.2 Autoregressive (AR) Model
533(1)
20.2.3.3 Autoregressive Moving Average (ARMA) Model
534(1)
20.2.3.4 The Spectra of MA, AR and ARMA Processes
535(1)
20.2.3.5 Second-Order Autoregressive [ AR(2)] Model
537(1)
20.2.3.6 Estimating the Spectrum
538(1)
20.2.4 Estimating the Spectrum Using Multiple EEG Traces
539(1)
20.2.4.1 Other Averaged Estimators
539(1)
20.2.4.2 Estimating Power in Specific Frequency Bands
540(1)
20.2.4.3 Detecting Outliers
540(1)
20.2.5 Confidence Intervals
540(1)
20.3 Spectral Analysis of Multichannel EEG
541(10)
20.3.1 Fourier-Domain Approach
542(1)
20.3.1.1 The Fourier-Cramer Representation
542(1)
20.3.1.2 The Spectral Matrix of an EEG
542(1)
20.3.1.3 Non-Parametric Estimator of the Spectral Matrix
544(1)
20.3.2 Time-Domain Approach
544(1)
20.3.3 Estimating Partial Coherence
545(3)
20.3.4 Estimating the Spectral Matrix Using Multiple EEG Traces
548(1)
20.3.4.1 Estimating the Spectral Matrix in Specific Frequency Bands
549(1)
20.3.5 Modeling and Inference on Connectivity
549(1)
20.3.5.1 Granger Causality
550(1)
20.3.5.2 Partial Directed Coherence (PDC)
550(1)
20.3.5.3 Summary of Metrics for Connectivity
551(1)
20.4 Spectral Analysis for High-Dimensional Data
551(4)
20.4.1 Methods for Fitting VAR Model on Multivariate Time Series
552(1)
20.4.1.1 Least Squares Estimation
552(1)
20.4.1.2 LASSO
552(1)
20.4.1.3 LASSLS
553(1)
20.4.2 EEG Data Analysis via LASSLS Methods
554(1)
20.4.2.1 VAR Modeling on High-Dimensional Multichannel EEG
554(1)
20.4.2.2 Inference on Directed Connectivity
555(1)
20.5 Source Localization and Estimation
555(7)
20.5.1 Overview of Source Models for EEG Data
555(1)
20.5.1.1 Dipole Source Model
556(1)
20.5.1.2 Independent Source Model
558(1)
20.5.1.3 A Generalized Model of EEG Signals
559(1)
20.5.2 Inverse Source Reconstruction
559(1)
20.5.2.1 Parametric Methods
559(1)
20.5.2.2 Imaging Methods
560(1)
20.5.2.3 Summary
562(1)
Bibliography
562(5)
21 Advanced Topics for Modeling Electroencephalograms
567(60)
Hernando Ombao
Anna Louise Schroder
Carolina Eucin
Chee-Ming Ting
Balqis Samdin
21.1 Introduction
568(4)
21.2 Clustering of EEGs
572(8)
21.2.1 Proposal: The Spectral Merger Clustering Method
572(1)
21.2.1.1 Total Variation Distance
572(1)
21.2.1.2 Hierarchical Spectral Merger Algorithm
573(1)
21.2.2 Analysis of Epileptic Seizure EEG Data
574(6)
21.3 Change-Point Detection
580(8)
21.3.1 Existing Methods and Challenges
580(2)
21.3.2 The FreSpeD Method
582(1)
21.3.2.1 Comparison to the Other Approaches
584(1)
21.3.3 Analysis of the Multichannel Seizure EEG Data
585(1)
21.3.3.1 Seizure Localization
585(1)
21.3.3.2 Seizure Onset Estimation and Potential Precursors
585(3)
21.4 Modeling Time-Varying Connectivity Using Switching Vector Autoregressive Models
588(11)
21.4.1 Background on Vector Autoregressive (VAR) Models
589(1)
21.4.1.1 Stationary VAR Model
589(1)
21.4.1.2 Time-Varying VAR Model
590(1)
21.4.1.3 Switching VAR (SVAR) Model
591(1)
21.4.2 Parameter Estimation
592(1)
21.4.3 Estimating Dynamic Connectivity States in Epileptic EEG
593(6)
21.5 Best Signal Representation for Non-Stationary EEGs
599(9)
21.5.1 Overview of Signal Representations
599(1)
21.5.2 Overview of SLEX Analysis
600(3)
21.5.3 Selecting the Best SLEX Signal Representation
603(3)
21.5.4 SLEX Analysis of Multichannel Seizure EEG
606(2)
21.6 Dual-Frequency Coherence Analysis
608(10)
21.6.1 Overview and Historical Development
610(2)
21.6.2 The Local Dual-Frequency Cross-Periodogram
612(1)
21.6.3 Formalizing the Concept of Evolutionary Dual-Frequency Spectra
612(1)
21.6.3.1 Harmonizable Process: Discretized Frequencies
612(1)
21.6.3.2 A New Model: The Time-Dependent Harmonizable Process
613(1)
21.6.3.3 Dual-Frequency Coherence between Bands
613(1)
21.6.4 Inference on Local Dual Frequency Coherence
614(1)
21.6.5 Local Dual Frequency Coherence Analysis of EEG Data
615(1)
21.6.5.1 Description of the Data and Experiment
615(1)
21.6.5.2 Implementation Details
615(1)
21.6.5.3 Results and Discussion
616(1)
21.6.6 Conclusion
617(1)
21.7 Summary
618(3)
Bibliography
621(6)
Index 627
Hernando Ombao is Professor in the Department of Statistics at the University of California, Irvine and Fellow of the American Statistical Association. Martin Lindquist is Professor in the Department of Biostatistics at Johns Hopkins University and Fellow of the American Statistical Association. Wesley Thompson is Associate Professor in the Department of Psychiatry at the University of California, San Diego and Lead Scientist at the Institute of Biological Psychiatry, Mental Health Services, Copenhagen, Denmark. John Aston is Professor in the Statistical Laboratory at the University of Cambridge and Fellow of the American Statistical Association.