Hi Donna,
I reran the error correction with the segmented volume loaded in Volume 2.
There are still 9 cavities and 1 segmentation object and the 1 remaining
error was not corrected. In fact, the automatic error correction did not
appear to correct any errors since '_corr' was not appended to the file name
and even the object list remained the same. Why didn't error correction
occur?
Also, what is the difference between a cavity and a handle? Which of these
errors is worse?
-Mellanie
- Original Message -
From: Donna Hanlon [EMAIL PROTECTED]
To: Caret, SureFit, and SuMS software users
[EMAIL PROTECTED]
Sent: Tuesday, July 20, 2004 2:26 PM
Subject: Re: [caret-users] (no subject)
Hi Mellanie,
Your segmentation has nine cavities and after moving the crosshairs well
within the red part of the segmentation and floodfilling, update handle
count reports 4 handles.
So, I'd recommend re-running automatic error correction in your case.
It may fill the cavities and will definitely make sure you don't have
multiple segmentation objects. It might even fix some handles
introduced by patching. At the very least, you'll get an updated Locate
Objects listing.
One other thing I forgot to mention the last time around: The -x
checkbox on the cropping tab should not have been checked. These volume
are named like *_pad*, which means padding has been added along the
cut face. Cropping the contralateral hemisphere isn't considered a cut
face; full hemispheres (like this one) should have no cut face boxes
checked.
I'm not sure what trouble this might cause in caret, but SureFit
probably won't copy border files to your SURFACES directory. This is
less important now that caret5 has template borders of its own.
Donna
On 07/20/2004 01:10 PM, [EMAIL PROTECTED] wrote:
Hi Donna,
I have uploaded the segmentation file:
nov16_2002A.R.full.segment_pad_vent_corr.patchtest17.mnc
The sMRI.mnc file has not changed.
Thanks,
Mellanie
Quoting Donna Hanlon [EMAIL PROTECTED]:
Hi Mellanie,
Great news, and I sure needed right now (lots of system problems this
week).
Go ahead and upload your segmentation volume here:
http://pulvinar.wustl.edu/cgi-bin/upload.cgi
If your sMRI.mnc or .params changed from before, also upload those
volumes.
It isn't unusual to have a handle so tiny you can't see it on the
surface and can't see any crossovers when you bring it into caret.
Handles like that almost never cause trouble during flattening or
registration. But I'll make sure this handle falls into that category.
Donna
On 07/20/2004 12:14 PM, [EMAIL PROTECTED] wrote:
Hi Donna,
Thanks to your great advice there is only 1 handle left on the surface.
However, I am having some difficulty finding where the error lies.
Would
you
be able to please take a look at the image to see whether you can
provide me
with some clues about the location of the offending voxels?
Thanks,
Mellanie
___
caret-users mailing list
[EMAIL PROTECTED]
http://pulvinar.wustl.edu/mailman/listinfo/caret-users
___
caret-users mailing list
[EMAIL PROTECTED]
http://pulvinar.wustl.edu/mailman/listinfo/caret-users
___
caret-users mailing list
[EMAIL PROTECTED]
http://pulvinar.wustl.edu/mailman/listinfo/caret-users
___
caret-users mailing list
[EMAIL PROTECTED]
http://pulvinar.wustl.edu/mailman/listinfo/caret-users
