MRI Analysis Syllabus

Aim

• Be able to use the main FSL tools to perform standard analyses of structural, functional and diffusion imaging data.
• Have a basic theoretical understanding of the underlying principles used by the FSL software.
• Be aware of the advantages and disadvantages of different alternative approaches to certain analyses and be able to make appropriate choices of methods and parameters.
• Have a good understanding of the GLM and how to set-up individual and group designs to answer specific application questions.
• Be able to write a methods section of an FMRI paper suitable for a journal publication.
  

Syllabus

Registration and unwarping

• Linear and non-linear registration (FLIRT and FNIRT)
• Discussion of atlases and standard spaces
• Practical and theoretical introduction to unwarping (EPI distortion / FUGUE)

Segmentation and Atrophy

• Basic tissue-type segmentation (FAST)
• Shape-based segmentation and modelling (FIRST and FreeSurfer)
• Structural change analysis - VBM-style morphometry
• Principles of atrophy detection in SIENA

Basic statistics

• Probability and hypothesis testing
• T-tests
• Z-statistics

Advanced Experimental Design

• Beyond block and event-related designs
• Sampling and bias
• Sparse sampling and special cases

FMRI Pre-processing

• Motion correction
• Slice timing corrections
• Spatial and Temporal Filtering

FMRI single subject analysis - basic

• GLM
• t and F contrasts
• Multiple comparison problem
• Pre-whitening
  

FMRI group analysis

• Random/mixed effects
• All-in-one model vs hierarchical
• Modeling variances
• FLAME/Bayes

FMRI analysis - advanced

• Basis functions
• Orthogonalisation
• Interactions
• Confounds
  

FMRI project

• Analysing a real group study and writing it up in a journal style
• First level analysis - exploring pre-stats choices, different design options, basis functions and HRFs.
• Group modeling - FLAME/OLS, fixed/mixed effects comparison, different variance groups.
  

FMRI exploratory analysis

• PCA and ICA theory (without detailed matrix maths)
• Practical guide - subjectivity, denoising with MELODIC
  

Diffusion: Tensors and TBSS

• Reconstruction of tensors
• FA/ADC calculation
• Quantitative differences
• TBSS (registration of tracts and measurement of change)

Diffusion: Tractography

• Tensor model and alternative models (e.g. cherry-on-stick)
• Tractography
• Crossing fibres

Assessment

• Answers to selected practical exercises (as indicated in the instruction sheets)
  
• FMRI project report on individual and group FMRI analysis
  
• Completion of the feedback form

Recommended Reading

• Huettel, Song and McCarthy (2009) Functional Magnetic Resonance Imaging (Chapter 8: Preprocessing of fMRI Data; and Chapter 10: Statistical Analysis)
• Jezzard, Smith and Matthews (2003) Functional Magnetic Resonance Imaging: An Introduction to Methods (Chapters 11, 12 and 15)
• FSL documentation, technical reports and papers
• Glossary of MRI Analysis terms
  

Course Organisation