In your case roughly 4GB of RAM.
Internally, probtrackx will use six times 50 volumes, each volume
being 256x256x31*sizeof(float).
(+ more if you specify waypoints/targets etc.)
Saad.
ps: if you can avoid upsampling the original data, that would save you
a lot of RAM.
On 4 Nov 2009, at 13:26, Yonghui Li wrote:
> Dear FSL,
>
> I am sure that I set the input directory for probtrackx correctly,
> which should be the output directory of BEDPOSTX (im my case, it is
> bedpostx.bedpostX). So, can I assume that the suggestion of Saad is
> the solution for my problem? More RAM will do the job? How many RAM
> is required to run probtrackx? The dimension of my 4D data is
> 256*256*45*31, voxel size 1*1*3.
>
> Best
> Yonghui
>
>
> Hi Yonghui Li
> This is a RAM problem. You need to run probtrackx on a computer with
> more RAM.
>
> Cheers,
> Saad.
>
>
>
>
> > Hi Yonghui,
>
> >
>
> > The the folder bedpostx.bedpostX seems fine so far.
>
> >
>
> > However when running Probtrackx you need to select the
> bedpostx.bedpostX -
>
> > folder, not the BEDPOSTX folder, as you want to use the OUTPUT from
>
> > bedpostx, not the input in order to execute Probtrackx.
>
> >
>
> > Does this help?
>
> >
>
> > Cheers,
>
> > Markus
>
> >
>
> >
>
> > 2009/11/4 YonghuiLi <[log in to unmask]>
>
> >
>
> > > Dear Markus
>
> > >
>
> > > a) I run the "probtrackx" through the GUI window. I first set
> the BEDPOSTX
>
> > > directory, then I set the X Y Z coordinate to one single voxel I
> am
>
> > > interested in, then I "GO".
>
> > >
>
> > > b) I set the name of input directory for "bedpostx" step is
> "bedpostx", so
>
> > > after it is done, the output directory is "bedpostx.bedpostX".
> When doing
>
> > > "ls bedpostx.bedpostX", it shows the following ifnormation:
>
> > > bvals mean_f2samples.nii.gz
> merged_ph1samples.nii.gz
>
> > > bvecs mean_ph1samples.nii.gz
> merged_ph2samples.nii.gz
>
> > > commands.txt mean_ph2samples.nii.gz
> merged_th1samples.nii.gz
>
> > > dyads1.nii.gz mean_th1samples.nii.gz
> merged_th2samples.nii.gz
>
> > > dyads2.nii.gz mean_th2samples.nii.gz monitor
>
> > > logs merged_f1samples.nii.gz
> nodif_brain_mask.nii.gz
>
> > > mean_f1samples.nii.gz merged_f2samples.nii.gz xfms
>
> > >
>
> > > Thank you very much for your help!
>
> > > Best
>
> > > Yonghui
>
> > >
>
> > >
>
> > > ======== 2009-11-03 18:14:06 您在来信中写道: ========
>
> > >
>
> > >
>
> > > Hi!
>
> > >
>
> > > In order to help, we need to know more. Can you post
>
> > > a) the probtrackx command you run?
>
> > > b) what you get when doing "ls <your_bedpostx.bedpost>"
>
> > >
>
> > > It is most likely that there is a file missing or has the wrong
> name or the
>
> > > wrong path...
>
> > >
>
> > > Hth,
>
> > > Markus
>
> > >
>
> > >
>
> > > 2009/11/3 YonghuiLi <[log in to unmask]>
>
> > >
>
> > >> Dear FSL,
>
> > >>
>
> > >> I run into some troubles while try to do the probtrackx on
> single voxel.
>
> > >> The "eddycurrent" and "bedpostx" steps are completed. Then I
> continued the
>
> > >> "probtrackx" by enter the coordinate of a single seed voxel.
> Then "GO". But
>
> > >> it stopped very soon, showing an error message:
>
> > >> "Errors: Image Exception: #99 :: Out of memory
>
> > >> terminate called after throwing an instance of
>
> > >> 'RBD_COMMON::BaseException'
>
> > >> Then the whole procedure stopped.
>
> > >>
>
> > >> Your help to my situation here will be highly appreciated.
>
> > >>
>
> > >> All the bests!
>
> > >>
>
> > >> Sincerely,
>
> > >> Yonghui Li
>
> > >>
>
> > >> Email: [log in to unmask]
>
> > >> 2009-11-03
>
> > >>
>
> > >
>
> > >
>
> > >
>
> > > --
>
> > >
>
> > >
>
> > > Dr. med. Markus Gschwind, M.D.
>
> > > Laboratory for Neurology and Imaging of Cognition - Dept of
> Neurosciences
>
> > > University Medical Center (CMU)
>
> > > 1, Michel-Servet
>
> > > CH-1211 GENEVA - Switzerland
>
> > >
>
> > > Tel 0041 (0) 22 379 5324
>
> > > Fax 0041 (0) 22 379 5402
>
> > > email: [log in to unmask]
>
> > > http://labnic.unige.ch
>
> > >
>
> > > = = = = = = = = = = = = = = = = = = = = = =
>
> > >
>
> > > 致
>
> > > 礼!
>
> > >
>
> > > *Yonghui Li,* PhD, PostDoc
>
> > >
>
> > > LIAMA Center for Computational Medicine (CMC)
>
> > > National Laboratory of Pattern Recognition (NLPR)
>
> > > Institute of Automation, Chinese Academy of Sciences (CASIA)
>
> > > 95 Zhong Guan Cun East Road, Hai Dian District, Beijing 100190,
> P.R.China
>
> > > Tel: +86-10-62659278
>
> > > Fax: +86-10-62551993
>
> > >
>
> > > National Reserach Center for Intelligent Computing Systems (NCIC)
>
> > > Institute of Computing Technology, Chinese Academy of Sciences
> (ICT)
>
> > > 6 Ke Xue Yuan South Road, Hai Dian District, Beijing 100190,
> P.R.China
>
> > >
>
> > > Email: [log in to unmask] or [log in to unmask] <[log in to unmask]
> >
>
> > > [log in to unmask]
>
> > > 2009-11-04
>
> > >
>
> > >
>
> >
>
> >
>
> >
>
> > --
>
> >
>
> >
>
> > Dr. med. Markus Gschwind, M.D.
>
> > Laboratory for Neurology and Imaging of Cognition - Dept of
> Neurosciences
>
> > University Medical Center (CMU)
>
> > 1, Michel-Servet
>
> > CH-1211 GENEVA - Switzerland
>
> >
>
> > Tel 0041 (0) 22 379 5324
>
> > Fax 0041 (0) 22 379 5402
>
> > email: [log in to unmask]
>
> > http://labnic.unige.ch
>
> >
>
>
>
>
--
Saad Jbabdi
University of Oxford, FMRIB Centre
JR Hospital, Headington, OX3 9DU, UK
+44 (0) 1865 222523 (fax 717)
www.fmrib.ox.ac.uk/~saad
|