Setup ICA Defaults
The explanation of hidden user interface controls is as follows:
- Select Type Of Data Pre-processing - Data is pre-processed prior to the
first data reduction. Options are discussed below.
- 'Remove Mean Per Timepoint' - At each time point, image mean is removed.
- 'Remove Mean Per Voxel' - Time-series mean is removed at each voxel.
- 'Intensity Normalization' - At each voxel, time-series is scaled to have a
mean of 100. Since the data is already scaled to percent signal change, there is
no need to scale the components.
- 'Variance Normalization' - At each voxel, time-series is linearly detrended
and converted to Z-scores.
- What Mask Do You Want to Use?
- Default Mask
- Mask is calculated using all the files for subjects and sessions or
only the first file for each subject and session depending upon the
variable DEFAULT_MASK_OPTION value in defaults. Boolean AND operation is
done to include the voxels that surpass the mean of each subject's
session.
- Select Mask
- Masks must be in Analyze or Nifti format.
- Select Type Of PCA
- There are three options like 'Standard' , 'Expectation Maximization' and
'SVD'.
Optional parameters for PCA are provided. Please see
PCA options
page.
- Select The Back-Reconstruction Type
- Options available are GICA, Regular (GICA2), GICA3 and Spatial-temporal
regression. GICA2 and GICA3 are not not shown in the GUI but can be called
in the batch script. GICA is a more robust tool to back reconstruct
components when compared to GICA2 and GICA3 for low model order.
- GICA - GICA uses PCA whitening and dewhitening information to back
reconstruct the component maps and timecourses ([1]).
- Spatial-temporal Regression - Back reconstruction is done using a two step
multiple regression ([3]). In the first step,
aggregate component spatial maps are used as basis functions and projected on to
the subject's data resulting in subject component time courses. In the second
step, subject component time courses are used as basis functions and projected
on to the subject's data resulting in component spatial maps for that subject.
-
Note:
- GICA, GICA2 timecourses are similar to the timecourses obtained using
Spatial-temporal Regression.
- Spatial maps obtained using GICA2 are exactly equal to the GICA3 method.
- All the back reconstruction methods give the same spatial maps and
timecourses for one single subject single session analysis.
- GICA and Spatial-temporal Regression component timecourses are
equivalent when 100% variance is retained in the first step PCA.
- Do You Want To Scale Components?
- Scale To Original Data(%)
- Scales components to represent percent signal change. Explanation of
calibrating components to percent signal change is given in
scaling components section.
- Z-scores
- Component images and time courses are divided by their standard
deviation.
- Scaling in Timecourses
- Spatial maps are normalized using the maximum intensity value and
the maximum intensity value is multiplied to the timecourses.
- Scaling in Maps and Timecourses
- Spatial maps are scaled using the standard deviation of timecourses
and timecourses are scaled using the maximum spatial intensity value.
- Note: By default, subject component images are centered
based on the peak of the distribution. Please see variable CENTER_IMAGES in
"icatb_defaults.m".
- How Many Data Reduction (PCA) Steps Do You Want to Run?
- The number of times you want to do PCA.
- Note: The number of data
reduction
steps depends on the number of data sets. A maximum of three data reduction
steps is allowed.
- Number of PC to reduce each group into
- Each subject is reduced to the number of principal components selected.
- Number of times the prompt strings requesting to enter the PC is
equal to the number of data reduction steps.
- Note:
- Always choose a greater number for PC in the first step if you wish
to choose different numbers for the principal components.
- The number of principle components after the last step of data
reduction (will be disabled as you already entered the number of
independent components) is the same as the number of independent
components you want to extract.
Figure 1: Setup Defaults menu