Run Analysis
Run Analysis contains initializing parameters along with the steps involved
in the group ICA. When you click on Run Analysis button, a figure
window will open to select the ICA parameter file. After selecting the ICA
parameter file, figure 1 will open and contains the following options:
- All*** - You can run all the group analysis steps at once or run each one
separately. To run all steps at once select 'All***', to run each step
separately select the steps in order.
- Parameter Initialization - All the variables that are needed later on or
during the analysis are declared and initialized. A parameter error check is
also preformed to try and catch errors before the group analysis begins.
- Data Reduction - Each data-set is reduced using Principal Components
Analysis (PCA). These reduced data-sets are then concatenated into a group or
groups depending on the number of data reductions steps selected, this process
is repeated. Each reduced data is saved in a MAT file and will be used in the
back reconstruction step. The second data reduction step demands most computing
resources (RAM) but works well for most current desktop/laptop PC configurations
and MATLAB versions. If an out of memory error from MATLAB occurs, try the usual
steps to improve performance, e.g. clear the workspace, terminate unused
programs and background processes. If it still does not work, reduce the number
of principal components or change the input data (e.g. lower sampling rate).
- Calculate ICA - The concatenated data from the data reduction step is used
and the aggregate ICA components are saved in MAT file format. Each component
consists of EEG time domain signal and weights.
- Back Reconstruction
- Regular - The aggregate components and the results from data reduction are
used to compute the individual subject components.
- Spatial-temporal Regression - The aggregate components and the original
data are used to compute the individual subject components.
- Calibrating Components - The original EEG data is used to scale the
components from arbitrary units to data units or Z-scores.
- Group Stats - The individual back reconstructed components are used to
compute statistics on components like mean, standard deviation and t-test. Each
component consists of EEG time domain signal and topographic weights. For the
t-test weights of a session used is the same as the mean of weights of that
session. These group stats components are calculated for each session and are
saved in MAT file format.
Results during each of the steps are printed to the MATLAB command window.
Before the group analysis can take place the parameters file must be
initialized. At this stage variables that were entered through the GUI are
parsed and initialized as needed for the group ICA analysis. Error checking is
also performed so that mistakes can be caught before the analysis starts. After
the analysis is completed, Display GUI will open automatically for
visualizing components. You can turn off this option by setting variable value
of OPEN_DISPLAY_GUI to 0 in 'icatb_defaults.m' file.
Note:
- The parameter initialization, data reduction, calculate ICA, back
reconstruction and group stats steps are the same as in GIFT.
- All the analysis information is stored in the '_results.log' file. This
file gets appended each time you run the analysis with the same prefix for the
output files.
- Output files information and variables contained in MAT files are explained
in Appendix C.
Figure 1: Run Analysis