This book describes methods for statistical brain imaging data analysis from both the perspective of methodology and from the standpoint of application for software implementation in neuroscience research. These include those both commonly used (traditional established) and state of the art methods. The former is easier to do due to the availability of appropriate software. To understand the methods it is necessary to have some mathematical knowledge which is explained in the book with the help of figures and descriptions of the theory behind the software. In addition, the book includes numerical examples to guide readers on the working of existing popular software. The use of mathematics is reduced and simplified for non-experts using established methods, which also helps in avoiding mistakes in application and interpretation. Finally, the book enables the reader to understand and conceptualize the overall flow of brain imaging data analysis, particularly for statisticians and data-scientists unfamiliar with this area.
The state of the art method described in the book has a multivariate approach developed by the authors team. Since brain imaging data, generally, has a highly correlated and complex structure with large amounts of data, categorized into big data, the multivariate approach can be used as dimension reduction by following the application of statistical methods. The R package for most of the methods described is provided in the book. Understanding the background theory is helpful in implementing the software for original and creative applications and for an unbiased interpretation of the output. The book also explains new methods in a conceptual manner. These methodologies and packages are commonly applied in life science data analysis. Advanced methods to obtain novel insights are introduced, thereby encouraging the development of new methods and applications for research into medicine as a neuroscience.
| Preface |
|
iv | |
|
|
|
1 | (9) |
|
|
|
3 | (7) |
|
|
|
10 | (21) |
|
|
|
10 | (4) |
|
Structural brain imaging data |
|
|
11 | (1) |
|
Functional brain imaging data |
|
|
12 | (2) |
|
|
|
14 | (1) |
|
|
|
15 | (16) |
|
|
|
16 | (6) |
|
|
|
22 | (2) |
|
|
|
24 | (5) |
|
|
|
29 | (2) |
|
3 Common Statistical Approach |
|
|
31 | (39) |
|
Structural brain imaging analysis |
|
|
31 | (1) |
|
|
|
31 | (3) |
|
|
|
34 | (1) |
|
Multiple comparison and correction |
|
|
35 | (8) |
|
|
|
43 | (1) |
|
|
|
44 | (10) |
|
|
|
45 | (4) |
|
Simple example for cluster size test |
|
|
49 | (3) |
|
|
|
52 | (2) |
|
|
|
54 | (5) |
|
|
|
56 | (3) |
|
|
|
59 | (6) |
|
|
|
61 | (4) |
|
Permutation based multiple correction |
|
|
65 | (5) |
|
|
|
65 | (5) |
|
|
|
70 | (81) |
|
|
|
70 | (4) |
|
|
|
71 | (3) |
|
|
|
74 | (22) |
|
Principal component analysis |
|
|
74 | (1) |
|
|
|
75 | (2) |
|
Two-steps dimension reduction |
|
|
77 | (4) |
|
|
|
81 | (15) |
|
|
|
96 | (3) |
|
|
|
97 | (2) |
|
|
|
99 | (5) |
|
|
|
102 | (2) |
|
Classification method and prediction model |
|
|
104 | (30) |
|
|
|
105 | (1) |
|
|
|
106 | (3) |
|
|
|
109 | (1) |
|
|
|
110 | (4) |
|
|
|
114 | (1) |
|
|
|
115 | (2) |
|
|
|
117 | (1) |
|
Logistic regression model |
|
|
117 | (3) |
|
|
|
120 | (3) |
|
|
|
123 | (4) |
|
|
|
127 | (7) |
|
|
|
134 | (13) |
|
|
|
134 | (2) |
|
|
|
136 | (2) |
|
|
|
138 | (9) |
|
|
|
147 | (3) |
|
|
|
148 | (2) |
|
|
|
150 | (1) |
|
|
|
151 | (43) |
|
|
|
151 | (24) |
|
|
|
151 | (2) |
|
|
|
153 | (5) |
|
|
|
158 | (1) |
|
|
|
159 | (1) |
|
|
|
159 | (4) |
|
|
|
163 | (1) |
|
|
|
164 | (7) |
|
Independent component analysis for fMRI |
|
|
171 | (4) |
|
|
|
175 | (10) |
|
|
|
176 | (1) |
|
|
|
177 | (4) |
|
|
|
181 | (2) |
|
|
|
183 | (2) |
|
|
|
185 | (1) |
|
|
|
185 | (9) |
|
|
|
186 | (1) |
|
Coordinate based meta-analysis (CBMA) |
|
|
187 | (2) |
|
|
|
189 | (3) |
|
|
|
192 | (2) |
| References |
|
194 | (17) |
| Index |
|
211 | |
Dr. Atsushi Kawaguchi is a professor in the Faculty of Medicine, Saga University in Japan. He received a Ph.D. in Mathematical Statistics from Kyushu University. He has conducted original research and published articles on brain imaging data analysis. He has produced collaborative works with medical doctors as a biostatistician. He has contributed chapters in books on (Frontiers of Biostatistical Methods and Applications in Clinical Oncology, Statistical Techniques for Neuroscientists, etc.). He is an associate editor of Journal of Statistics, and Journal of Biometrics, both in Japanese. He received the Biometric Society of Japan Encouragement Prize in 2010.
More info about Taylor & Francis e-books