About head motion in DPARSFA

Submitted by Louise on

YAN Chao-Gan

Thu, 11/28/2013 - 04:43

Permalink

Hi,

 
Framewise displacement (FD) is a summary motion metric combining the 6 translation and rotation parameters. As such, it's not equivalent to that setting a single threshold for each of the 6 parameters (e.g., 1mm and 1 degree).
 
In my point of view, I am not very comfortable for the cost of scrubbing -- temporal discontinuity (as for ALFF and fALFF), as well as for the variety of degrees of freedom across subjects. However, at the current stage, I will perform the analysis both without and with scrubbing, and at least put the latter into supplementary analysis.
 
Best,
 
Chao-Gan

jtang

Thu, 11/28/2013 - 09:30

Permalink

.
DPARSFA  进行数据处理，进行fmri数据的标准化，选用了DARTEL 进行标准化，采用了T1数据，出现系列报错，请问是如何解决？ 电脑是64位 win7， 8G内存。
如果不用DARTEL的话，可以顺利运行。

谢谢！

...............................................................................................................................................??? Error using ==> single
Out of memory. Type HELP MEMORY for your options.

Error in ==> file_array.subsasgn>subfun at 127
dat   = feval(dt(ind).conv,dat);

Error in ==> file_array.subsasgn at 70
        subfun(sobj,double(dat),args{:});

Error in ==> y_Write4DNIfTI at 65
dat(:,:,:,:)=Data;

Error in ==> y_RegressOutImgCovariates at 155
y_Write4DNIfTI(VolumeAfterRemoveCov,Header_Out,[OutputDir,filesep,'CovRegressed_4DVolume.nii']);

Error in ==> DPARSFA_run>makeF%26/F% at 2119
            y_RegressOutImgCovariates([AutoDataProcessParameter.DataProcessDir,filesep,FunSessionPrefixSet{iFunSession},AutoDataProcessParameter.StartingDirName,filesep,AutoDataProcessParameter.Subje
            Error in ==> parallel_function at 454
            F(base, limit);

Error in ==> DPARSFA_run at 2012
        parfor i=1:AutoDataProcessParameter.SubjectNum

Error in ==> DPARSFA>pushbuttonRun_Callback at 1601
    [Error]=DPARSFA_run(handles.Cfg);

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

Error in ==> DPARSFA at 33
    gui_mainfcn(gui_State, varargin{:});

??? Error while evaluating uicontrol Callback


On Thu, Nov 28, 2013 at 12:43 PM, 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 YAN Chao-Gan (YAN Chao-Gan)

Hi,

 
Framewise displacement (FD) is a summary motion metric combining the 6 translation and rotation parameters. As such, it's not equivalent to that setting a single threshold for each of the 6 parameters (e.g., 1mm and 1 degree).
 
In my point of view, I am not very comfortable for the cost of scrubbing -- temporal discontinuity (as for ALFF and fALFF), as well as for the variety of degrees of freedom across subjects. However, at the current stage, I will perform the analysis both without and with scrubbing, and at least put the latter into supplementary analysis.
 
Best,
 
Chao-Gan

Online version of this post: http://rfmri.org/comment/2974#comment-2974





--
Jinsong Tang, MD
Institute of Mental Health
The Sec

YAN Chao-Gan

Thu, 11/28/2013 - 12:20

Permalink

Hi,
From the message, this happens after DARTEL, at the step of nuisance covariates regression.

You can try restrating the system and only run matlab,  redo the analysis just start from this step.

To check available memory, you can type  

feature('memstats')

in matlab command to check the biggest memory block you can use. 
Free Blocks:       
1. [at 10007000]   1546 MB   (60a69000)

BTW, What's the voxel size and number of time points of your dataset?

Best,

Chao-Gan



On Thu, Nov 28, 2013 at 4:30 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 Jtang (jtang)
.
DPARSFA  进行数据处理，进行fmri数据的标准化，选用了DARTEL 进行标准化，采用了T1数据，出现系列报错，请问是如何解决？ 电脑是64位 win7， 8G内存。
如果不用DARTEL的话，可以顺利运行。

谢谢！

...............................................................................................................................................??? Error using ==> single
Out of memory. Type HELP MEMORY for your options.

Error in ==> file_array.subsasgn>subfun at 127
dat   = feval(dt(ind).conv,dat);

Error in ==> file_array.subsasgn at 70
        subfun(sobj,double(dat),args{:});

Error in ==> y_Write4DNIfTI at 65
dat(:,:,:,:)=Data;

Error in ==> y_RegressOutImgCovariates at 155
y_Write4DNIfTI(VolumeAfterRemoveCov,Header_Out,[OutputDir,filesep,'CovRegressed_4DVolume.nii']);

Error in ==> DPARSFA_run>makeF%26/F% at 2119
            y_RegressOutImgCovariates([AutoDataProcessParameter.DataProcessDir,filesep,FunSessionPrefixSet{iFunSession},AutoDataProcessParameter.StartingDirName,filesep,AutoDataProcessParameter.Subje
            Error in ==> parallel_function at 454
            F(base, limit);

Error in ==> DPARSFA_run at 2012
        parfor i=1:AutoDataProcessParameter.SubjectNum

Error in ==> DPARSFA>pushbuttonRun_Callback at 1601
    [Error]=DPARSFA_run(handles.Cfg);

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

Error in ==> DPARSFA at 33
    gui_mainfcn(gui_State, varargin{:});

??? Error while evaluating uicontrol Callback


On Thu, Nov 28, 2013 at 12:43 PM, 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 YAN Chao-Gan (YAN Chao-Gan)

Hi,

 
Framewise displacement (FD) is a summary motion metric combining the 6 translation and rotation parameters. As such, it's not equivalent to that setting a single threshold for each of the 6 parameters (e.g., 1mm and 1 degree).
 
In my point of view, I am not very comfortable for the cost of scrubbing -- temporal discontinuity (as for ALFF and fALFF), as well as for the variety of degrees of freedom across subjects. However, at the current stage, I will perform the analysis both without and with scrubbing, and at least put the latter into supplementary analysis.
 
Best,
 
Chao-Gan

Online version of this post: http://rfmri.org/comment/2974#comment-2974





--
Jinsong Tang, MD
Institute of Mental Health
The Sec
Online version of this post: http://www.rfmri.org/comment/2975#comment-2975





--
Chao-Gan YAN, Ph.D.

jtang

Thu, 11/28/2013 - 12:40

Permalink

Hi Chao-Gan
Thanks a lot!  I will try it.

The  voxel size = 3*3*3, and the time point =250. 

Best,

Jinsong

On Thu, Nov 28, 2013 at 8:20 PM, 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 YAN Chao-Gan (YAN Chao-Gan)
Hi,
From the message, this happens after DARTEL, at the step of nuisance covariates regression.

You can try restrating the system and only run matlab,  redo the analysis just start from this step.

To check available memory, you can type  

feature('memstats')

in matlab command to check the biggest memory block you can use. 
Free Blocks:       
1. [at 10007000]   1546 MB   (60a69000)

BTW, What's the voxel size and number of time points of your dataset?

Best,

Chao-Gan



On Thu, Nov 28, 2013 at 4:30 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 Jtang (jtang)
.
DPARSFA  进行数据处理，进行fmri数据的标准化，选用了DARTEL 进行标准化，采用了T1数据，出现系列报错，请问是如何解决？ 电脑是64位 win7， 8G内存。
如果不用DARTEL的话，可以顺利运行。

谢谢！

...............................................................................................................................................??? Error using ==> single
Out of memory. Type HELP MEMORY for your options.

Error in ==> file_array.subsasgn>subfun at 127
dat   = feval(dt(ind).conv,dat);

Error in ==> file_array.subsasgn at 70
        subfun(sobj,double(dat),args{:});

Error in ==> y_Write4DNIfTI at 65
dat(:,:,:,:)=Data;

Error in ==> y_RegressOutImgCovariates at 155
y_Write4DNIfTI(VolumeAfterRemoveCov,Header_Out,[OutputDir,filesep,'CovRegressed_4DVolume.nii']);

Error in ==> DPARSFA_run>makeF%26/F% at 2119
            y_RegressOutImgCovariates([AutoDataProcessParameter.DataProcessDir,filesep,FunSessionPrefixSet{iFunSession},AutoDataProcessParameter.StartingDirName,filesep,AutoDataProcessParameter.Subje
            Error in ==> parallel_function at 454
            F(base, limit);

Error in ==> DPARSFA_run at 2012
        parfor i=1:AutoDataProcessParameter.SubjectNum

Error in ==> DPARSFA>pushbuttonRun_Callback at 1601
    [Error]=DPARSFA_run(handles.Cfg);

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

Error in ==> DPARSFA at 33
    gui_mainfcn(gui_State, varargin{:});

??? Error while evaluating uicontrol Callback


On Thu, Nov 28, 2013 at 12:43 PM, 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 YAN Chao-Gan (YAN Chao-Gan)

Hi,

 
Framewise displacement (FD) is a summary motion metric combining the 6 translation and rotation parameters. As such, it's not equivalent to that setting a single threshold for each of the 6 parameters (e.g., 1mm and 1 degree).
 
In my point of view, I am not very comfortable for the cost of scrubbing -- temporal discontinuity (as for ALFF and fALFF), as well as for the variety of degrees of freedom across subjects. However, at the current stage, I will perform the analysis both without and with scrubbing, and at least put the latter into supplementary analysis.
 
Best,
 
Chao-Gan

Online version of this post: http://rfmri.org/comment/2974#comment-2974





--
Jinsong Tang, MD
Institute of Mental Health
The Sec
Online version of this post: http://www.rfmri.org/comment/2975#comment-2975





--
Chao-Gan YAN, Ph.D.Online version of this post: http://www.rfmri.org/comment/2977#comment-2977





--
Jinsong Tang, MD
Institute of Mental Healt

Louise

Mon, 12/09/2013 - 13:05

Permalink

Hi,

Chao-Gan

I didn't cheak my blog in time in the past days. I feel really sorry for that. Thanks for your useful suggestion! I think I may do some further calculation.

Best,

Louise

 

Hi,

I have a question when dealing with the head motion using DPARSFA. Actually, I have performed 'scrubbing' with my data.  If the head motion of the volume exceeding 0.5mm (FD-Jenkinson), then I deleted one time point before and two time points after this volume. 

Then my question is how to clarify the scrubbing method and the creteria more standard in the menu. Do I have to show the translation and rotation? And what is the translation and rotation when using the cretira of FD-Jenkinson>0.5mm? 

And what's your view about scrubbing? Is this method accpted more widely now?   

Thank you very much and best wishes!

Louise