Hi there,
> 24 hours does seem a bit excessive. It usually takes rather less time
> than this, but the time does depend on the image dimensions, number of
> iterations, CPU etc. I can't actually think of what may cause it to
> have major problems. Maybe it has finished now?
One would certainly hope so since the original message was posted 2 days
ago :) But just as a hunch: have you coregistered the images to each
other before warping them? This may be a good idea as a rigid-body will
take care of gross positional differences and will leave only the small
local differences for the warping to model. If I remember correctly,
this was recommended for earlier versions of the toolbox so perhaps this
helps.
Best,
Marko
> -----Original Message-----
> From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]]
> On Behalf Of Hedok Lee
> Sent: Saturday, August 25, 2007 12:38 AM
> To: [log in to unmask]
> Subject: [SPM] Time it takes to finish high-dimensional warping
>
> Dear SPMers:
>
> This is regarding high-dimensional warping in SPM5(the latest patch
> Updates
> 826). I'm trying to register serially corrected T1W scans using high-
> dimensional warping. It's taken from the same subject using high
> resolution T1W(SPGR 1.0x1.0x1.5mm 256x256x100). I'm interested in the
> anatomical areas where structures have expanded or shrunk between the
> two
> time points, so I thought it would be useful to try the toolbox.
>
> This is the first time trying this toolbox, and my apology for such a
> elementary question, and please correct me if this is not something I
> should be doing. I picked high-dimensional warping in the GUI menu, and
>
> selected image from time point A as a reference. In the "moved image",
> I
> selected an image taken in time point B. Everything else was set to
> default values.
>
> The problem is that it hasn't finished the registration after ~24hrs and
>
> I'm wondering if this is a normal behavior. I'm using a PC(AMD64 1.8GHz
> 2G
> RAM Centos 4_64) with Matlab 7.1.0.183
>
> According to the "top" command, matlab is using 14% of RAM and none of
> the
> swap space is being utilized so I don't think it's RAM problem. The CPU
>
> has been 100% the whole time. I appreciate if someone can tell me if
> it's
> going to take days, weeks, or years to register two scans. Also I'm
> attaching the screenshot of what I've been seeing for past 24hrs.
>
> Thanks,
>
> Hedok
>
>
--
=====================================================================
Marko Wilke (Dr.med./M.D.)
[log in to unmask]
Universitäts-Kinderklinik University Children's Hospital
Abt. III (Neuropädiatrie) Dept. III (Pediatric neurology)
Hoppe-Seyler-Str. 1, D - 72076 Tübingen
Tel.: (+49) 07071 29-83416 Fax: (+49) 07071 29-5473
=====================================================================
|