erro messages when extracting ROI times series and functional connectivity

Submitted by carloscat on Tue, 02/02/2010 - 20:27

Hi:

I used DPARSF for extracting ROI time series and functional connectivity on previously precprocessed data.
My data are 53x63x46, so I've reslice mask with Coreg-Reslice under SPM as the video mentioned.
But error messages showed that wrong mask dimension was noted.
Moreover, these error messages remained even I chose "null mask" item.
Could someone tell me how to fix this problem?
(I've pasted the messages below)

Thanks

??? Error using ==> rest_loadmask at 64

    Mask does not match.
    Mask size is 61x73x61, not same with required size 53x63x46

Error in ==> rest_Y_ExtractROITC at 48
                maskROI =rest_loadmask(nDim1, nDim2, nDim3, AROIDef{x});

Error in ==> DPARSF_run at 728
        rest_Y_ExtractROITC([AutoDataProcessParameter.DataProcessDir,filesep,FunImgDir,filesep,AutoDataProcessParameter.SubjectID{i}],
        CovariatesROI,[AutoDataProcessParameter.DataProcessDir,filesep,F
Error in ==> DPARSF>pushbuttonRun_Callback at 924
    [Error]=DPARSF_run(handles.Cfg);

Error in ==> gui_mainfcn at 96
        feval(varargin{:});

Error in ==> DPARSF at 38
    gui_mainfcn(gui_State, varargin{:});

??? Error while evaluating uicontrol Callback

Submitted by carloscat on Wed, 02/03/2010 - 15:39

Hi:

I've found that I made a mistake. I only reslice BrainMask file instead of all Mask files (CSF, white matter).
But further error messages were noted when extracting ROI time series.
Did I miss something?

Thanks

ROI time courses retrieving through "ROI Center(mm)=(0, -50, 30); Radius=5.00 mm.".??? Attempted to access mask(26,20.3333,27.6667); index must be a positive integer or logical.

Error in ==> rest_SphereROI>Ball2Mask at 265
            mask(x, y, z) =1;

Error in ==> rest_SphereROI at 110
    theMask =Ball2Mask(ABrainSize, AVoxelSize, AROICenter, AROIRadius, Header.Origin);

Error in ==> rest_Y_ExtractROITC at 36
            maskROI =rest_SphereROI( 'BallDefinition2Mask' , AROIDef{x}, BrainSize, VoxelSize, Header);

Error in ==> DPARSF_run at 864
        rest_Y_ExtractROITC([AutoDataProcessParameter.DataProcessDir,filesep,FunImgDir,filesep,AutoDataProcessParameter.SubjectID{i}],
        AutoDataProcessParameter.CalFC.ROIDef,[AutoDataProcessParameter.
Error in ==> DPARSF>pushbuttonRun_Callback at 920
    [Error]=DPARSF_run(handles.Cfg);

Error in ==> gui_mainfcn at 96
        feval(varargin{:});

Error in ==> DPARSF at 37
    gui_mainfcn(gui_State, varargin{:});

??? Error while evaluating uicontrol Callback

Submitted by YAN Chao-Gan on Thu, 02/04/2010 - 23:38

Hi!
It's a little weird.
What's the detailed information of your data? Or could you paste several data here? I can give more suggestions if I can repeat your error.
Best wishes!
                  Chao-Gan

Submitted by carloscat on Fri, 02/05/2010 - 17:28

Hi:

I really appreciate your efforts and thanks for your  feedback.
My data were acquired under Siemens 3T system, with TR=2.5, slice number=40 with interleaved order, and 200 timepoints.
I did not use DPARSF for preprocessing and used our own protocol for our data, ex.: with higher interpolation.
But after that I arrange data to the directories as you presented (FunImg, FunImgNormalized, FunImgNormalizedSmoothed...).
The data could now finished further process including detrend and filtering but failed when extracting ROI time series.
Or should I send my preprocessed data to you?

Thanks

Carlos

Submitted by YAN Chao-Gan on Fri, 02/05/2010 - 21:36

Hi!
I think this problem is caused by your own protocol.
It needs the information of the origin of your data when extracting ROI time courses, usually the NIfTI format will not concur any problem.
Does your own protocol make ANALYZE format images other than NIfTI format images? Then you need to pay attention to the information of origin.
Best wishes!
              Chao-Gan

Submitted by carloscat on Sat, 02/20/2010 - 22:29

Hi:

Happy Chinese New Year!
It seems that the file format should not be the origin of the error since I performed all my preprocessed under spm8 as your video showed.
But I chose unwarp during realignment and also used highest(7th) interpolation for my data.
I've tried to use REST step by step instead of automatically processing with DPARSF.
But the error messages still be the same when running extracting time seires of ROI.
So I'm not sure why this still happened.

Thanks for your hard working and helping me to find the problems!

Carlos

Submitted by YAN Chao-Gan on Thu, 02/25/2010 - 16:08

Hi!
Did you mean you want to extract time series based on data without normalization?
Then REST will face problems since the voxel size is not integer.
Best wishes!

Submitted by carloscat on Tue, 03/02/2010 - 14:15

No. I have done all the preprocessing steps including normalization.
But both of REST or DPARSF would encounter above error messages for extracting time courses.
I still have no idea about what's wrong with this since the only difference is using higher interpolaration and unwarp in preprocessing.
I temporally used MARSBAR for extracting time courses but still want to understand where the problems are.

Thanks for your reply!
Best Regards

Submitted by YAN Chao-Gan on Wed, 03/03/2010 - 23:30

Hi!
Could you send me several image files with higher interpolaration and without higher interpolaration?
Paste here as attachments or send to my email: ycg.yan#gmail.com is OK.
Best wishes!

Submitted by pingfan2008 on Fri, 06/14/2013 - 19:43

这个情况，我也遇到了，得去dparsf_run.mat文件里找到那三个模板，之后把模板的参数改成我们自己数据的参数。

具体改这个：
  CovariatesROI=[];
    if (AutoDataProcessParameter.Covremove.WholeBrain==1)
        CovariatesROI=[CovariatesROI;{[ProgramPath,filesep,'Templates',filesep,'BrainMask_05_61x73x61.img']}];
    end
    if (AutoDataProcessParameter.Covremove.CSF==1)
        CovariatesROI=[CovariatesROI;{[ProgramPath,filesep,'Templates',filesep,'CsfMask_07_61x73x61.img']}];
    end
    if (AutoDataProcessParameter.Covremove.WhiteMatter==1)
        CovariatesROI=[CovariatesROI;{[ProgramPath,filesep,'Templates',filesep,'WhiteMask_09_61x73x61.img']}];
    end

不知新版的dparsf这个地方改进了没？