234
Comment:
|
3277
|
Deletions are marked like this. | Additions are marked like this. |
Line 1: | Line 1: |
= Principles of FMRI data analysis = | = Principles of analysis = The aim of these pages is to introduce the theory behind different processing stages. If you're interested in the practicalities of how to do an analysis, you might want the [:AnalysisSoftware:software] page instead. |
Line 3: | Line 4: |
* SliceTiming * PrinciplesSmoothing |
Below is an outline list of the areas of theory involved in functional imaging, and related documents that may help. Unfortunately some of these sources are difficult, but in my experience, all of them repay a slow reread, even if you think you understood almost nothing first time round. There is a bias towards SPM, but the same principles usually apply to other imaging packages. = Introduction to imaging = == Basic checks == * Check for CommonArtefacts * Run DataDiagnostics - this is done by most [:aa:AutomaticAnalysisIntroduction] recipes if you choose to use this path * If you are unfortunate enough to find bad data, consider whether to RemoveBadScans == Typical pre-processing stages == These are the stages typical to most analyses * [:SliceTiming:Slice timing] - correcting for the fact that different slices in the brain are acquired at different times * [:FmBackground:Undistort EPIs using field maps] - usually no longer necessary on data from the Siemens scanner * [:MotionCorrection:Motion correction] - correcting for subject movement through an fMRI scan, sometimes called realignment * PrinciplesSpatialProcessing * ProcessingCoregistration * ProcessingNormalization * ProcessingSmoothing == Statistics == |
Line 6: | Line 23: |
* [:BasicOnsetFiles:Basic steps to making onset files] | |
Line 7: | Line 25: |
* PrinciplesRandomEffects * SmallVolumeCorrection * RegionAnalysis * [:MultiVoxelPatternAnalysis:Multi voxel pattern analysis (MVPA)] = Introductions to the theory and algorithms behind SPM = == Presentations == * CBU SpmMiniCourse2007 * Rik Henson's SpmMiniCourse2008 * [javascript:void(0);/*1268925552787*/ Introduction to fMRI methods 2010] == Written guides == * [http://www.fil.ion.ucl.ac.uk/spm/ FIL/SPM home page] * SpmQuickStart * The [http://www.fil.ion.ucl.ac.uk/spm/doc/books/hbf2/ Human Brain Function book] - these are a series of chapters covering theory and to some extent implementation in SPM of the areas below. Be warned that some of these chapters are aimed at quite a high level, and may be hard to follow for those without relevant background. * [http://www.fil.ion.ucl.ac.uk/spm/dox.html Papers published by the FIL group on SPM] - these are available online via the SPM website, as are some other important papers on related theory. * [http://www.fil.ion.ucl.ac.uk/spm/course/ SPM course notes] = More advanced issues = * ParametricModulations - and the issue of orthogonalisation * DealingWithDifference - HRF responses can differ from area to area * ["Templates"] * FmriAnalysisProblems |
|
Line 8: | Line 46: |
* Templates * MniTalairach |
* MniTalairach - differences between the MNI brain used by SPM and the Talairach atlas |
Line 12: | Line 49: |
* AnatomyLinks = Useful background = * ImagingJournals * A very basic [http://imaging.mrc-cbu.cam.ac.uk/pdfs/matrices.pdf primer on matrices], in pdf format (see the [:ReadingOnlineDocs:online reading page] if you need help with pdf documents). |
Principles of analysis
The aim of these pages is to introduce the theory behind different processing stages. If you're interested in the practicalities of how to do an analysis, you might want the [:AnalysisSoftware:software] page instead.
Below is an outline list of the areas of theory involved in functional imaging, and related documents that may help. Unfortunately some of these sources are difficult, but in my experience, all of them repay a slow reread, even if you think you understood almost nothing first time round. There is a bias towards SPM, but the same principles usually apply to other imaging packages.
Introduction to imaging
Basic checks
Check for CommonArtefacts
Run DataDiagnostics - this is done by most [:aa:AutomaticAnalysisIntroduction] recipes if you choose to use this path
If you are unfortunate enough to find bad data, consider whether to RemoveBadScans
Typical pre-processing stages
These are the stages typical to most analyses
[:SliceTiming:Slice timing] - correcting for the fact that different slices in the brain are acquired at different times
[:FmBackground:Undistort EPIs using field maps] - usually no longer necessary on data from the Siemens scanner
[:MotionCorrection:Motion correction] - correcting for subject movement through an fMRI scan, sometimes called realignment
Statistics
[:BasicOnsetFiles:Basic steps to making onset files]
[:MultiVoxelPatternAnalysis:Multi voxel pattern analysis (MVPA)]
Introductions to the theory and algorithms behind SPM
Presentations
Rik Henson's SpmMiniCourse2008
- [javascript:void(0);/*1268925552787*/ Introduction to fMRI methods 2010]
Written guides
[http://www.fil.ion.ucl.ac.uk/spm/ FIL/SPM home page]
The [http://www.fil.ion.ucl.ac.uk/spm/doc/books/hbf2/ Human Brain Function book] - these are a series of chapters covering theory and to some extent implementation in SPM of the areas below. Be warned that some of these chapters are aimed at quite a high level, and may be hard to follow for those without relevant background.
[http://www.fil.ion.ucl.ac.uk/spm/dox.html Papers published by the FIL group on SPM] - these are available online via the SPM website, as are some other important papers on related theory.
[http://www.fil.ion.ucl.ac.uk/spm/course/ SPM course notes]
More advanced issues
ParametricModulations - and the issue of orthogonalisation
DealingWithDifference - HRF responses can differ from area to area
- ["Templates"]
MniTalairach - differences between the MNI brain used by SPM and the Talairach atlas
Useful background
A very basic [http://imaging.mrc-cbu.cam.ac.uk/pdfs/matrices.pdf primer on matrices], in pdf format (see the [:ReadingOnlineDocs:online reading page] if you need help with pdf documents).