[Mrtrix-discussion] streamtrack- how to reduce calculation time

[Donald Tournier]

Hi Vinod,

Almost 6 millions tracks generated! No wonder that took a while...

Apart from running on a faster computer (i.e. multi-core), the simplest way
to get the tracking to run faster is to use a more targeted seed ROI, to
increase the proportion of tracks that actually reach the intended
destination. If you're already using a very focused ROI, then maybe the fact
that you are getting 1 in 5000 tracks to reach your target ROI is telling
you that the tract is a very low probability one... Maybe your seed region
is a bit off?

And running on a grid computer sounds like it's not going to work. The
software is multithreaded, and will run on a multi-core system, but not on a
distributed network cluster - it's a very different type of architecture...

Not sure if that helps...

Cheers!

Donald.


On Wed, May 6, 2009 at 3:39 PM, vinod kumar <mail.vinod at yahoo.com> wrote:

> Dear mrtrix users and donald,
>
> greetings :)
>
> It was nice to see you in hawaii :)
>
> I tried to do some new tractography, by including several -exclude ROIs.  (
> as follows)
>
> >streamtrack SD_PROB $orig2_dir/CSD12_sf06.mif -seed
> $orig2_dir/roi/roi1.nii -mask $orig2_dir/mask.mif
> $orig2_dir/results/roi1_track.tck -include $orig2_dir/roi/roi_2.nii -exclude
> $orig2_dir/roi/roi_3.nii -exclude $orig2_dir/roi/roi_4.nii
>
> > 5801844 generated,     1000 selected    [100%]
>
>
> this took almost a day. Is there some way I can reduce the calculation time
> ? or can run this command on sun grid ( parallel on different machines). We
> have sun grid engine for few software, but I don't know how to incroporate
> mrtrix in that.
>
> Thanx
>
> vinod
>
>
>
>
>
>
> _______________________________________________
> Mrtrix-discussion mailing list
> Mrtrix-discussion at www.nitrc.org
> http://www.nitrc.org/mailman/listinfo/mrtrix-discussion
>
>


-- 
Jacques-Donald Tournier (PhD)
Brain Research Institute, Melbourne, Australia
Tel: +61 (0)3 9496 4078
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.nitrc.org/pipermail/mrtrix-discussion/attachments/20090506/5b8269e7/attachment.html