We have two fMRI learning tracks; the fast track and the slow track. The fast fMRI track is useful for short term (1-2 months) interns and senior project students. For those who wish to get deeper into fMRI (Master’s & PhD students, postdocs, long-term interns) should check the other track.
Fast-track fMRI
In this track, you will become acquainted with concepts such as:
- fMRI pre-processing on a basic scale
- General linear model
- Univariate analysis (beta values, contrasts, t-maps, f-maps and such)
- Basic fMRI experiment design
- Preparation of event/onset files using MATLAB or Python
- SPM (Statistical Parametrical Mapping) toolbox
That being said, here is a list of materials which can be covered within 10 days that will bring you to a position to do basic fMRI analysis:
- fMRI theory
- Brief introduction to neuroimaging
- fMRI Bootcamp -MITCBMM (watch the first 3 videos)
- Univariate analysis
- HRF Overview
- Introduction to GLM
- Additional GLM material
- SPM 1st level analysis
- Preprocessing theory playlist by Monti UCLA
- fMRI analysis playlist (Particularly the GLM part to understand GLM better)
- Event/onset files
- Statistics
- FIR analysis
- Contents of SPM.mat file
- Alternative to above
Slow-track fMRI
This track covers all relevant features of fMRI including the physics, pre-processing steps, file and database formats, analysis methods, visualization of analysis results, fMRI experiment design efficiency.
- Brief introduction to neuroimaging
- fMRI Physics
- Introduction to NIfTI format
- BIDS format
- Preprocessing
- Preprocessing theory playlist by Monti UCLA
- Andy’s brain blog pre-processing
- CBU Analysis principles (read for additional exposure to the concepts)
- CBU pre-processing primer (read for additional exposure to the concepts)
- Fieldmaps
- After completing the material up until this part, you can go to the “Fast-track fMRI” part and read the material there. Once you complete those sections, come back here and continue.
- Additional material (SPM course)
- MRIcroGL
- MRIcroGL videos by Andrew Jahn
- MRIcroGL introduction
- MRIcroGL manual
- Basis functions
- Parametric modulation
- Design efficiency
- Data diagnostics
- Once the material above is completed, you will know univariate analysis. Then you can move onto more advanced analysis methods.
Advanced fMRI analysis methods
- MVPA – Machine Learning classification
- RSA – Representational Similarity Analysis
- Representational similarity analysis – connecting the branches of systems neuroscience
- Representational Similarity Analyses: A Practical Guide for Functional MRI Applications (accessible via Bilkent)
- A Guide to Representational Similarity Analysis for Social Neuroscience
- A toolbox for representational similarity analysis
- Linear discriminant t analysis (LD-t)
- Data diagnostics
- PPI as GLM covariates
- Bayesian inference for neuroimaging
MRI Physics & acquisition
- How MRI works: https://www.youtube.com/playlist?list=PLkSVzqeK5v2C4X1G3IuRUQeNveNGNxZrn
- MR Physics: https://www.youtube.com/playlist?list=PLBkfUPj1TSRqmh2pXV2t5bY5Qr3Z8AeOi
- MRI Pulse sequences playlist
- Information regarding protocol parameters
- MRI Toolbox (for selecting protocol parameters)
- MRC CBU Imaging Sequences
- Tips for data acquisition
- Understanding fMRI artifacts: https://practicalfmri.blogspot.com/2011/11/understanding-fmri-atifacts.html
- How to understand good data: https://practicalfmri.blogspot.com/2011/11/understanding-fmri-artifacts-good.html
- Setting your experiment not for success, but less failure
- Harvard fMRI physics and protocol Q&A: https://cbs.fas.harvard.edu/science/core-facilities/neuroimaging/information-investigators/MRphysicsfaq
- 3T User Training FAQ (by UC Berkeley Dr. Ben Inglis): Download
Extras