Diff for "SpmAnalysis" - Meg Wiki
location: Diff for "SpmAnalysis"
Differences between revisions 1 and 50 (spanning 49 versions)
Revision 1 as of 2007-04-14 13:45:51
Size: 781
Editor: RikHenson
Comment:
Revision 50 as of 2009-01-26 20:12:12
Size: 3313
Editor: RikHenson
Comment:
Deletions are marked like this. Additions are marked like this.
Line 2: Line 2:
SPM5 can perform basic preprocessing of MEG data, and source reconstruction, including ECD and variants of L2-norm imaging solutions. It's weaknesses are: i) ECD is limited (e.g, the Neuromag Xfit program is more flexible), ii) it's graphics are slow. It's strengths are: i) well integrated with MRI and fMRI analysis (eg, segmentation), ii) in Matlab, so relatively easy to understand and extend, iii) includes advanced statistical methods (e.g, for multiple comparisons across space or time), and novel variants of the L2-norm approach to the inverse problem, iv) includes group-based analysis (eg in MNI/Talairach space).
Line 4: Line 3:
For an introduction, see these slides: RiksSlides Many of us here use SPM for EEG and/or MEG analysis. The main reason is that SPM is freeware (requiring only Matlab), and thus modifiable and relatively easily to understand (if you have basic knowledge of a procedural computer language, given that Matlab is a fairly high-level language). Several of us are actively extending SPM for our own purposes, particularly in relation to the Neuromag type of MEG data.
Line 6: Line 5:
First, you will need to convert your *.FIF files into SPM files. ----
 * For specific demo using data from our Neuromag MEG machine, see SpmDemo
 * For a fuller demo of other EEG/MEG analysis in SPM5 (though from a different MEG machine), including more general features (e.g, time-freq analysis, 3D statistical maps), with proper step-by-step instructions via the GUI, see: http://www.fil.ion.ucl.ac.uk/spm/data/mmfaces.html
 * For a more theoretical introduction to source localisation in SPM5, see these slides: attachment:henson-SPM-Grad08-4-meeg.ppt
----
Here are some relevant papers:

 * Summary of localisation approach using ReML for evoked and induced responses (mathematical; cites earlier development papers too): attachment:FristonEtAl_hbm_06.pdf
 * Basic considerations for Group Analyses (though using individual meshes): attachment:HensonEtAl_NI_07.pdf
 * Use of inverse-normalised ("canonical") cortical meshes: attachment:MattoutEtAl_JCIN_07.pdf
 * Choice of forward models for MEG (e.g, single-sphere vs BEM), including further validation of canonical meshes: attachment:HensonEtAl_NI_inpress.pdf
 * '''New method of Multiple Sparse Priors (MSP): attachment:FristonEtAl_NI_08_MSP.pdf'''
----
Additional SPM5 functions:

 * If you are not using the version of SPM5 installed on the CBU network (ie do not have */spm5/cbu_updates on your Matlab path) - eg are an external collaborator - you will need to download the latest versions of some site-specific SPM5 functions in order to read the FIF format data files from our MEG scanner. These SPM functions can be downloaded from here: http://imaging.mrc-cbu.cam.ac.uk/svn/spm5_cbu_updates/devel (see also http://www.mrc-cbu.cam.ac.uk/~rh01/fif2spm.html).

 * Miscellaneous helpful SPM/Matlab functions we've written can be found here: http://imaging.mrc-cbu.cam.ac.uk/svn/meg_misc/devel/ For permission to contribute to or edit this repository, contact Ian Nimmo-Smith.

 * A meta-toolbox of functions that allow you to apply EEGLAB functions (eg. ICA, plot raw data, plot ERFimage) to SPM-format MEG data can be found here: http://imaging.mrc-cbu.cam.ac.uk/svn/spmeeglab/devel/ For further information, contact Jason Taylor. Help on artefact correction using ICA is forthcoming...

 * This page describes the procedure for creating a 3D SensorSpm (topography x time).

----
General note:

 * SPM5 can read raw and averaged FIF files, though you will probably first want to run your raw data through the [:Maxfilter:Maxfilter utility], particularly if you 1) used Active Shielding during acquisition, 2) if you want to apply (temporal) SSS to remove noise, 3) if you used continuous HPI and/or 4) if you want to transform all subjects to a common (device) space. Max Filter can also downsample (eg from 1000Hz to 200Hz) and convert the data into different datatypes (e.g, short), which will help reduce filesize and processing time.

Analysis of MEG Data in SPM5

Many of us here use SPM for EEG and/or MEG analysis. The main reason is that SPM is freeware (requiring only Matlab), and thus modifiable and relatively easily to understand (if you have basic knowledge of a procedural computer language, given that Matlab is a fairly high-level language). Several of us are actively extending SPM for our own purposes, particularly in relation to the Neuromag type of MEG data.


  • For specific demo using data from our Neuromag MEG machine, see SpmDemo

  • For a fuller demo of other EEG/MEG analysis in SPM5 (though from a different MEG machine), including more general features (e.g, time-freq analysis, 3D statistical maps), with proper step-by-step instructions via the GUI, see: http://www.fil.ion.ucl.ac.uk/spm/data/mmfaces.html

  • For a more theoretical introduction to source localisation in SPM5, see these slides: attachment:henson-SPM-Grad08-4-meeg.ppt


Here are some relevant papers:

  • Summary of localisation approach using ReML for evoked and induced responses (mathematical; cites earlier development papers too): attachment:FristonEtAl_hbm_06.pdf

  • Basic considerations for Group Analyses (though using individual meshes): attachment:HensonEtAl_NI_07.pdf

  • Use of inverse-normalised ("canonical") cortical meshes: attachment:MattoutEtAl_JCIN_07.pdf

  • Choice of forward models for MEG (e.g, single-sphere vs BEM), including further validation of canonical meshes: attachment:HensonEtAl_NI_inpress.pdf

  • New method of Multiple Sparse Priors (MSP): attachment:FristonEtAl_NI_08_MSP.pdf


Additional SPM5 functions:


General note:

  • SPM5 can read raw and averaged FIF files, though you will probably first want to run your raw data through the [:Maxfilter:Maxfilter utility], particularly if you 1) used Active Shielding during acquisition, 2) if you want to apply (temporal) SSS to remove noise, 3) if you used continuous HPI and/or 4) if you want to transform all subjects to a common (device) space. Max Filter can also downsample (eg from 1000Hz to 200Hz) and convert the data into different datatypes (e.g, short), which will help reduce filesize and processing time.

CbuMeg: SpmAnalysis (last edited 2013-03-08 10:02:27 by localhost)