The field of neuroimaging with functional magnetic resonance imaging (fMRI) is developing at a rapid pace, with a seemingly endless number of software packages, statistical methods, and different ways to organize and analyze neuroimaging data. Among such a wide variety of options, and with so many seemingly conflicting pieces of advice on the correct way of analyzing neuroimaging data, knowing what decisions to make is a difficult task.
Modern fMRI: Practical Lessons and Insights provides an up-to-date, holistic overview of the field of fMRI, familiarizing the reader with the latest trends in neuroimaging, such as standardized data organization and preprocessing, advances in functional connectivity and machine learning, and current guidelines in data and code sharing. This includes advice about best practices in preprocessing, statistical modeling, QA checks, and some of the latest tools and concepts to be familiar with, including fMRIPrep, OpenNeuro.org, Open Science practices, and Jupyter notebooks
1. Introduction: A brief history of neuroimaging and functional magnetic
resonance imaging
2. Acquisition parameters and your experiment: The intersection of scanning
protocols, experimental
design, and statistical power
3. Choosing your functional magnetic resonance imaging analysis software: An
introduction to the
big three (SPM, FSL, and AFNI), recent packages to be familiar with, and the
advantages of each
4. Choosing your programming language: Unix, MATLAB, Python, and the rise of
Jupyter Notebooks
5. Standardized data organization and preprocessing: The history and uses of
BIDS, fMRIPREP, and an
introduction to Neurodesk.org
6. Statistical modeling and correcting for multiple comparisons: The mass
univariate approach, recent developments, and what might work best for you
7. Region of interest analysis: The many ways to select and analyze a region,
and the strengths of each approach
8. Pitfalls of fMRI analysis: Circular analyses, biased ROIs, logical
fallacies, and how to avoid them
9. New developments in functional connectivity: Dynamic connectivity, graph
theory, and the connectome
10. New developments in machine learning: Representational similarity
analysis, hyperalignment, and their applications
11. Open science: An overview of preregistration, data sharing, and current
guidelines
12. Open-access databases, meta-analysis, and reproducibility
13. Bringing it all together: Summarizing the main points of this book
14. Where do we go from here? The future of neuroimaging analysis
Appendix A: Review of papers that question fMRI findingsWhat to learn from
them, and how to keep
them in perspective
Appendix B: AI and neuroimaging analysisHow Generative AI can inform the
preprocessing and analysis of fMRI data
Dr. Andrew Jahn is a research scientist in the Department of Radiology at the University of Michigan. He is the creator of Andys Brain Blog and its associated YouTube channel, online resources that host tutorials and videos about neuroimaging analysis from start to finish in all the major software packages. He continues to produce training materials and teach workshops about neuroimaging analysis, functional connectivity, machine learning, and other topics related to cognitive neuroscience.
