[Icatb-discuss] Errors for constrained ICA

[David Y. <kc...@gm...>]

Dear Experts,

Thanks very much for the excellent tool.

I am trying to perform constrained ICA (either Constrained ICA (Spatial)
or MOO-ICAR ) in our rsfMRI data (54 subjects with one session per subject)
using GroupICAT v4.0c.

I had two kinds of errors as below:

1. When applying “Spatial Reference Based ICA”

After generating the mask, I had the following errors:

Error using icatb_read_batch_file (line 420)

Select Number of IC to be more than or equal to 2.

Error in icatb_setup_analysis (line 13)

        varargout{1} = icatb_read_batch_file(inputFile);

Error in icatb_batch_file_run (line 60)

    param_file = icatb_setup_analysis(inputFiles{nFile});

Error in setup_reference_ica>generateBatch (line 722)

icatb_batch_file_run(batchFileName);

Error in setup_reference_ica>done_Callback (line 422)

generateBatch(handles);

Error in gui_mainfcn (line 95)

        feval(varargin{:});

Error in setup_reference_ica (line 42)

    gui_mainfcn(gui_State, varargin{:});

Error in
matlab.graphics.internal.figfile.FigFile/read>@(hObject,eventdata)setup_reference_ica('done_Callback',hObject,eventdata,guidata(hObject))



Error while evaluating UIControl Callback.

2. When applying "Standard ICA/IVA", I have different problems as could not
adequately include the subjects.

(1) When I keep each subject's data in one subject folder within the root
data folder, after loading the data, I have the following errors

Error using icatb_setup_analysis>set_data_para (line 384)
Please re-select the data as the no of selected files is found to be 1
(F:\002_DAT-FC-SCH\010_GIFT\Data\Smoothed_2\Subject001\srniftiNORMS_Subject001_Session001.nii,1)

Error in ==> set_data_para at 384




                                  Error in ==> dataCallback at 947

Similar problems occurred when I select each subject's data manually.

(2) When I put all subjects' data within the root data folder (no subject
folder), It seems ok that I can do the constrained ICA (no errors related
to the number of IC).
However, only one subject was included in the analysis, not the whole 54
subjects.

I wonder what will be possible ways to deal with these problems?

The  Preprocessing steps for my data included preprocessing by fmriprep
with default settings + denoising with default settings of the CONN
toolbox + resampled to 3x3x3  & smoothing by SPM12.

Thanks very much for the help.

Best,
KC

-- 
Kai-Chun Yang MD, PhD

Department of Psychiatry
Taipei Veterans General Hospital, Taipei, Taiwan