Error with Bet when processing two people at once but not when processing seperately

Submitted by David on

Hi,

I'm encountering a strange error with the bet procedure. That is, I tried to process two persons at once and then I get an error with bet when processing the second person. But what I don't understand is that if I process the two persons seperately, it all works fine for both persons. Here is the error I encounter:


0.739118 for pre-processing
0.140905 for normal computation
0.658801 for search along normal
0.052821 for checking self-intersection
0.517607 for output
Error using nifti/subsasgn>assigndat (line 367)

Unknown filename extension (.gz).

Error in nifti/subsasgn>fun (line 76)
            obj      = assigndat(obj,val);

Error in nifti/subsasgn (line 20)
        obji    = fun(obji,subs,val);

Error in y_Write (line 61)
NIfTIObject.dat=dat;

Error in y_Call_bet (line 47)
    y_Write(Data,Head,OutputFilename);

Error in DPARSFA_run>(parfor body) (line 1400)
            y_Call_bet(T1File, OutputFile_Temp, '');

Error in DPARSFA_run (line 1372)
        parfor i=1:AutoDataProcessParameter.SubjectNum

Error in DPARSFA>pushbuttonRun_Callback (line 1670)
    [Error]=DPARSFA_run(handles.Cfg);

Error in gui_mainfcn (line 96)
        feval(varargin{:});

Error in DPARSFA (line 30)
    gui_mainfcn(gui_State, varargin{:});

Caused by:
    Error using nifti/subsasgn>assigndat (line 367)
    Unknown filename extension (.gz).
 
Error while evaluating uicontrol Callback
 

Thanks in advance!

YAN Chao-Gan

Wed, 02/25/2015 - 17:02

Hi David,

From the error message, could you try to unzip the .nii.gz to .nii, and have a try again?
 
Best,
 
Chao-Gan

Hey,

after unzipping, Bet runs without error messages. But the normalisation is really bad and I will not be able to use the data.  Strangely when I process the same data seperateley the normalisation is fine. Also without Bet and running all datasets at once, normalisation mostly fails. But when processing sepereateley, normalisation is fine.

I ran some tests and I found out the following: I add two persons and process them at once. Then Bet (without unzipping) fails at the second person. When I add three persons and process them at once, Bet runs without error for the second person (where the error message was before), but gives the same error for the third person.

So there could be problem with the unzipping procedure in DPARSF?

I appreciate any hints, what the problem could be :)

Hi David,
Do you have multiple T1 images for a single subject? It's weird that the normalization effects is bad after unzipping, they should be the same. Are you running with windows? I will suggest you run it at another computer (e.g., linux) to see if you can find any hints. From the current message, I am not sure where the issues are.

Best,

Chao-Gan

On Sat, Feb 28, 2015 at 5:00 AM, The R-fMRI Network <rfmri.org@gmail.com> wrote:
[To post a comment, please reply to rfmri.org@gmail.com ABOVE this line]

Commented by David (David)

Hey,

after unzipping, Bet runs without error messages. But the normalisation is really bad and I will not be able to use the data.  Strangely when I process the same data seperateley the normalisation is fine. Also without Bet and running all datasets at once, normalisation mostly fails. But when processing sepereateley, normalisation is fine.

I ran some tests and I found out the following: I add two persons and process them at once. Then Bet (without unzipping) fails at the second person. When I add three persons and process them at once, Bet runs without error for the second person (where the error message was before), but gives the same error for the third person.

So there could be problem with the unzipping procedure in DPARSF?

I appreciate any hints, what the problem could be :)


Online version of this post: http://rfmri.org/comment/3555#comment-3555


Many a little makes a mickle -- your kind contributions shall make our efforts not perish from the earth. Please help The R-fMRI Network at http://rfmri.org/#overlay=HelpUs
To manage subscriptions, please visit: http://rnet.co/mailman/listinfo/rfmri.org_rnet.co
Mail comment ID: http://rfmri.org/mailcomment/redirect/%3C31.1958.3555.1425117599.c8f3974213bcdf09801bf888845eb74c%40www.rfmri.org%3E

Hey,

after trying it on a different computer, I think I found the problem. I used the parallel computing toolbox on a high performance system with 12 cores before. Using DPARSF with all cores somehow fails. I don't know why, but when trying it with only 4 cores it works fine, but not with 8,10 or 12 cores. Still don't know if that is really the problem, but at least the proprocessing works fine now :)

 

Thanks again

YAN Chao-Gan

Wed, 03/11/2015 - 21:20

In reply to by David

Hi David,
Thanks for your information! Yes, this is difficult to understand. Maybe some reading/writing conflict caused that problem.

Best,

Chao-Gan

On Tue, Mar 10, 2015 at 6:31 AM, The R-fMRI Network <rfmri.org@gmail.com> wrote:
[To post a comment, please reply to rfmri.org@gmail.com ABOVE this line]

Commented by David (David)

Hey,

after trying it on a different computer, I think I found the problem. I used the parallel computing toolbox on a high performance system with 12 cores before. Using DPARSF with all cores somehow fails. I don't know why, but when trying it with only 4 cores it works fine, but not with 8,10 or 12 cores. Still don't know if that is really the problem, but at least the proprocessing works fine now :)

 

Thanks again


Online version of this post: http://rfmri.org/comment/3572#comment-3572


Many a little makes a mickle -- your kind contributions shall make our efforts not perish from the earth. Please help The R-fMRI Network at http://rfmri.org/#overlay=HelpUs
To manage subscriptions, please visit: http://rnet.co/mailman/listinfo/rfmri.org_rnet.co
Mail comment ID: http://rfmri.org/mailcomment/redirect/%3C31.1958.3572.1425983461.0ccc984f55b055ba1b73bfbdb4241385%40www.rfmri.org%3E