Error using CompCor

Submitted by Emilia Cheladze on Fri, 01/13/2017 - 12:37
Hello!
I have ​resting state ​data acquired with ​the ​following scanning parameters: TR 2.3, vox size 3x3x3​,​​ and ​I am doing an analysis using DPRSFA V_2.1 and Matlab 2015b​. ​I am reslicing the voxel size ​to ​4x4x4 ​during ​the ​normaliz​ation step (this is so I can compare this data with another resting state data set at the group level)​. I am getting an error while doing the Nuisance Covariates Regression with CompCor and the step is not completed.​ Would anyone know what the problem is?​
 
Steps I did:
Remove first 10 time points
Realignment
Coregistration
Segmentation (Segment) 
Normalisation (using T1 Image unified segmentation; Voxel Size: 4x4x4)
Nuisance Covariates Regression (Rigid Body 6, method: CompCor)
 
I am getting​ the​ following error: 
 
Error using y_CompCor_PC (line 52)
Index exceeds matrix dimensions.
 
Error in DPARSFA_run (line 2457)
parfor i=1:AutoDataProcessParameter.SubjectNum
 
Error in DPARSFA>pushbuttonRun_Callback (line 1779)
[Error]=DPARSFA_run(handles.Cfg);
 
Error in gui_mainfcn (line 95)
feval(varargin{:});
 
Error in DPARSFA (line 30)
gui_mainfcn(gui_State, varargin{:});
 
Error while evaluating UIControl Callback
 
Thank you ​in advance ​for​ your​ help.
 
Best regards,
Emilia Cheladze

Forums:

YAN Chao-Gan

Fri, 02/03/2017 - 03:37

The nuisance masks are not the same voxel size with your data (4x4x4). Please reslice them first.

Emilia Cheladze

Tue, 02/07/2017 - 10:39

Thank you for the answer!

Could you please write, where can I find the masks and how (and at which point) can I reslice them?

Is the 3 3 3 voxel size the default one for the nuisance masks ? I didn't get the error when I used CompCor with 4 4 4 voxel size data sets after normalizing them with DARTEL - is the analysis incorrect then?

Best,

Emilia Cheladze

 

YAN Chao-Gan

Fri, 02/10/2017 - 09:01

Then just use DARTEL. Should be OK.