canlab_glm* are a set of functions for running and handling GLM analyses of fMRI data with a focus on batching (doing all subjects as easily as one). It provides a unified framework for doing GLM analyses which can give consistency to analyses across studies and also makes added functionality immediately available to all users. Ideally it will make possible anything you could do by doing SPM subject levels and robfit analyses directly, in addition to some commonly-needed functionality.

Some of the features
- automatic generation of subject level design review htmls and group level results htmls - dramatically faster via parallel processing (e.g., all subjects run at the same time) - email notification of successful/failed completion - on-the-fly session concatenation - allow for different number of runs per subject - allow for missing conditions in some subjects (e.g., error conditions) - option for conversion to single trials design - easy specification of conditions with wildcards - smart updating/overwriting behavior - appending/replacing contrasts in existing analyses

Primary functions
There are a lot of child functions and such in the repository (SCN_Core_Support/GLM_Batch_tools). The primary set of functions intended for direct use:
canlab_glm_subject_levels (runs subject levels with SPM)
canlab_glm_group_levels (runs group levels with robfit)
canlab_glm_publish (makes design review and results htmls)
canlab_glm_getinfo (retrieves descriptions of analyses (e.g., contrasts+weights, visualized design matrix, etc.))
canlab_glm_roistats (returns roi data and reports)

To learn about the functions:
First stop: help canlab_glm_subject_levels
This will describe that function, give you runtime options, and also tell you about the following two commands:
canlab_glm_subject_levels('README') will print an overview of canlab_glm functions.
canlab_glm_subject_levels('dsgninfo') will print a full description of the DSGN structure that fully defines the subject level design.

Although the dsgninfo provides examples of setting most of the parameters and options, in the repository you will also find SCN_Core_Support/GLM_Batch_tools/canlab_glm_example_DSGN_setup.txt, an example of the sort of script you can use to create a DSGN structure.

NOTE about canlab_glm_group_levels
Even if you don't use canlab_glm_subject_levels to do SPM subject level analyses, canlab_glm_group_levels can be handy for running group levels (it can often be run with defaults, will run)

help/fmri_help/canlab_glm.txt · Last modified: 2016/08/19 18:00 (external edit)
CC Attribution-Noncommercial-Share Alike 4.0 International
Driven by DokuWiki Recent changes RSS feed Valid CSS Valid XHTML 1.0