slicer image registration - scaling problem

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

slicer image registration - scaling problem

James Henry Zane Eastabrook
Hello all,

I'm having some difficulty registering some brain pets and mris.  The alignment and orientation are correct but my mr images have scaling problems in the axial and sagittal planes.  Is there a way to correct/adjust this within slicer or do I need to make the adjudment prior to loading the images?

Thanks
Zane

----- Original Message -----
From: [hidden email]
Date: Wednesday, February 3, 2010 10:41 pm
Subject: slicer-users Digest, Vol 59, Issue 2
To: [hidden email]

> Send slicer-users mailing list submissions to
> [hidden email]
>
> To subscribe or unsubscribe via the World Wide Web, visit
> http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users
> or, via email, send a message with subject or body 'help' to
> [hidden email]
>
> You can reach the person managing the list at
> [hidden email]
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of slicer-users digest..."
>
>
> Today's Topics:
>
>    1. Re: Slicer 3.4 - Rigid Registration (Andrew Wiles)
>    2. need help (Ahmed Shaibal)
>    3. Re: need help (Sylvain Jaume)
>    4. Use of Slicer2.6 as a workaround for Slicer3 3.5
> 3D volume      view (Madan Rao)
>    5. Re: Use of Slicer2.6 as a workaround for Slicer3
> 3.5 3D      volume view (Dominik Meier)
>    6. ImageLabelCombine Module (Justin Adams)
>    7. hide hidden folders in File dialogs (Luke Bloy)
>
>
> -----------------------------------------------------------------
> -----
>
> Message: 1
> Date: Mon, 01 Feb 2010 15:24:06 -0500
> From: "Andrew Wiles" <[hidden email]>
> Subject: Re: [slicer-users] Slicer 3.4 - Rigid Registration
> To: <[hidden email]>, "stella ang"
> <[hidden email]>Message-ID:
> <[hidden email]>Content-Type: text/plain;
> charset=US-ASCII
>
> Hello Stella,
>
> I am working on a Slicer Module that will collect fiducial
> points from a tracking system using openIGT link which can then
> be used with the Transform from Fiducials method. 
> Unfortunately, I am still chasing some memory leaks that I
> introduced and I haven't checked in the code yet.  I am
> also, converting my Linux distro from openSUSE to Ubuntu because
> of some silly configuration issues in openSUSE that make it very
> difficult to configure for Slicer and the other open source
> libraries (things just work on Ubuntu).  So if you can wait
> a few days, I hope to have new module checked in.
>
> Cheers.
> Andrew
>
> --------------------
> Andrew D. Wiles, PhD, PEng
> Research Scientist, Medical Division
> NDI
> 103 Randall Drive
> Waterloo, ON, Canada, N2V 1C5
> Telephone: +1 (519) 884-5142 ext. 314
> Toll Free: +1 (877) 634-6340
> Global: ++ (800) 634-634-00
> Facsimile: +1 (519) 884-5184
> Website: www.ndigital.com
> --------------------
>
> >>> stella ang <[hidden email]> 02/01/10 9:54 AM >>>
>
> Hi all,
>
> I wanted to perform a rigid registration. However, I met with
> some problems.
>
>
> I have linked up Aurora-Slicer3 (using OpenIGTLink) and now I
> intend to register the scanned CT file with the patient/body.
>
>
> What should be adjusted at the corresponding Data module? (e.g
> Locator node under Rigid Registration node?) Then what about the
> reference frame?
>
> If i'm not wrong, at Transform module, the Rigid Registration is
> supposed to response to probe movement, which will then conclude
> its success of registration? Am I right? But apparently I do not
> see such response at all. And what does "invert" means?
>
> Another note is sometimes the program will hang while I initiate
> these modules. Any particular reason?
>
> Thanks for your help.
>
> Regards,
> Stella
>
>                                               
> _________________________________________________________________
> Hotmail: Powerful Free email with security by Microsoft.
> https://signup.live.com/signup.aspx?id=60969
>
> --
>
>
>
>
> ------------------------------
>
> Message: 2
> Date: Tue, 2 Feb 2010 13:01:42 -0800 (PST)
> From: Ahmed Shaibal <[hidden email]>
> Subject: [slicer-users] need help
> To: [hidden email]
> Message-ID: <[hidden email]>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Hi
>
> I am a new user. I just finished a tutorial on 3D data loading
> and visualization. In this tutorial a DICOM format volume is
> loaded. The volume size is 256-by-256-by-124. That means it has
> 124 sagittal slices each having resolution 256-by-256. and the
> coronal and axial slices should?be 256-by-124. ?But from the 2D
> view it looks like that axial and coronal slices also have
> resolution 256-by-256. Seems like slicer does some kind of
> interpolation to resize those two slices. can anyone tell me
> what is that exactly or refer to any documentation?
>
> waiting for feedback.
> regards.?
> ?
> Ahmed Mostayed
> PhD Student
> Intelligent Systems for Medicine Lab (ISML)
> Dept of Mechanical Engineering
> The University of Western Australia
> 35 Stirling highway, Crawley-Perth, WA-6009
> Tel. +61430911286 Email. [hidden email]
>
>
>       New Email addresses available on Yahoo!
> Get the Email name you've always wanted on the new @ymail
> and @rocketmail.
> Hurry before someone else does!
> http://mail.promotions.yahoo.com/newdomains/aa/
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: http://massmail.spl.harvard.edu/pipermail/slicer-
> users/attachments/20100202/21e8e9e3/attachment-0001.html
>
> ------------------------------
>
> Message: 3
> Date: Tue, 2 Feb 2010 16:52:51 -0500
> From: Sylvain Jaume <[hidden email]>
> Subject: Re: [slicer-users] need help
> To: Ahmed Shaibal <[hidden email]>
> Cc: [hidden email]
> Message-ID:
> <[hidden email]>
> Content-Type: text/plain; charset=ISO-8859-1
>
> Hello Ahmed,
>
> Could you check the volume dimensions under Modules > Volumes >
> Info ?
> What index do you see in the upper left corner for each view?
>
> In case you are confused about the inter-voxel interpolation for
> rendering, you can uncheck the Interpolate checkbox under
> Modules >
> Volumes > Display.
>
> Regards,
> Sylvain
>
> On Tue, Feb 2, 2010 at 4:01 PM, Ahmed Shaibal
> <[hidden email]> wrote:
> > Hi
> >
> > I am a new user. I just finished a tutorial on 3D data loading and
> > visualization. In this tutorial a DICOM format volume is
> loaded. The volume
> > size is 256-by-256-by-124. That means it has 124 sagittal
> slices each having
> > resolution 256-by-256. and the coronal and axial slices should?be
> > 256-by-124. ?But from the 2D view it looks like that axial and
> coronal> slices also have resolution 256-by-256. Seems like
> slicer does some kind of
> > interpolation to resize those two slices. can anyone tell me
> what is that
> > exactly or refer to any documentation?
> >
> > waiting for feedback.
> > regards.
> >
> >
> >
> > Ahmed Mostayed
> > PhD Student
> > Intelligent Systems for Medicine Lab (ISML)
> > Dept of Mechanical Engineering
> > The University of Western Australia
> > 35 Stirling highway, Crawley-Perth, WA-6009
> > Tel. +61430911286 Email. [hidden email]
> > ________________________________
> > New Email addresses available on Yahoo!
> > Get the Email name you've always wanted on the new @ymail and
> @rocketmail.> Hurry before someone else does!
> > _______________________________________________
> > slicer-users mailing list
> > [hidden email]
> > http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users
> > To unsubscribe: send email to slicer-users-
> [hidden email]> with unsubscribe as the subject
> >
>
>
> ------------------------------
>
> Message: 4
> Date: Wed, 3 Feb 2010 19:49:36 +0530
> From: Madan Rao <[hidden email]>
> Subject: [slicer-users] Use of Slicer2.6 as a workaround for Slicer3
> 3.5 3D volume view
> To: [hidden email]
> Message-ID:
> <[hidden email]>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Hello,
>
> I am giving an example of a common problem in Slicer3 3.4/3.5
> (that I was
> facing) and its solution here.
>
> I have a dicom image set: 512 x  512 x 119, spacing:
> 0.46785, 0.46785, 1.5,
> Origin: -119.77, -119.77,-269.5.
>
> The 3D image is outside the 3D box in Conventional layout. This gives
> several problems for example in tumor segmentation
> or vascular modelling.
>
> If we change the values of image origin we are supposed to get a
> correctview. However, it is impractical since we do not know
> how to randomly change these values.
>
> While working on a solution I happen to find that
> Slicer2.6-opt-win32-x86-2006-08 not only can open 3D view
> correctly but also
> report any errors. In other words, Slicer2.6 can find the
> correct values of
> the Origin for us. When I tried to save the image as *.nhdr
> file, I got a memory overload error. I converted the dicom
> images to raw
> using mcverter and reloaded and saved as *.nhdr files.
>
> The correct values that load the 3Dview of the image are:
> 119.77, 119.77,
> 88.5.
>
> Hope this helps.
>
> AM Mohan Rao
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: http://massmail.spl.harvard.edu/pipermail/slicer-
> users/attachments/20100203/7ae86d63/attachment-0001.html
>
> ------------------------------
>
> Message: 5
> Date: Wed, 3 Feb 2010 10:15:12 -0500
> From: Dominik Meier <[hidden email]>
> Subject: Re: [slicer-users] Use of Slicer2.6 as a workaround for
> Slicer3 3.5 3D volume view
> To: Madan Rao <[hidden email]>
> Cc: [hidden email]
> Message-ID: <[hidden email]>
> Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed
>
> Since Slicer operates in mm space, the default center is the
> image 
> center (half the dimensions) times the voxel size, i.e.
> [0.46785, 0.46785, 1.5] .* [512 512 119] / 2 =  119.77,
> 119.77, 88.5.
>
> There is a "Center Volume" button in the "Volumes" module. You
> should 
> be able to change the Image Origin data directly.
>
> Dominik
>
> On Feb 3, 2010, at 9:19 AM, Madan Rao wrote:
>
> > Hello,
> >
> > I am giving an example of a common problem in Slicer3 3.4/3.5
> (that 
> > I was facing) and its solution here.
> >
> > I have a dicom image set: 512 x  512 x 119, spacing:
> 0.46785, 
> > 0.46785, 1.5, Origin: -119.77, -119.77,-269.5.
> >
> > The 3D image is outside the 3D box in Conventional layout.
> This 
> > gives several problems for example in tumor segmentation
> > or vascular modelling.
> >
> > If we change the values of image origin we are supposed to get
> a 
> > correct view. However, it is impractical since we do not know
> > how to randomly change these values.
> >
> > While working on a solution I happen to find that Slicer2.6-
> opt-
> > win32-x86-2006-08 not only can open 3D view correctly but also
> > report any errors. In other words, Slicer2.6 can find the
> correct 
> > values of the Origin for us. When I tried to save the image as
> *.nhdr> file, I got a memory overload error. I converted the
> dicom images 
> > to raw using mcverter and reloaded and saved as *.nhdr files.
> >
> > The correct values that load the 3Dview of the image are:
> 119.77, 
> > 119.77, 88.5.
> >
> > Hope this helps.
> >
> > AM Mohan Rao
> >
>
>
>
> ------------------------------
>
> Message: 6
> Date: Wed, 03 Feb 2010 13:06:11 -0500
> From: "Justin Adams" <[hidden email]>
> Subject: [slicer-users] ImageLabelCombine Module
> To: <[hidden email]>
> Message-ID: <[hidden email]>
> Content-Type: text/plain; charset=US-ASCII
>
>
> Question regarding the functioning of the ImageLabelCombine
> module.  I have been trying to merge two label maps
> together using this module (both are derived from the same
> original CT scan), but my resulting output label map is
> always... blank.  Since there is little in the entry for
> this module currently online, I was wondering if there were any
> specific instructions to using this feature. I am operating
> Slicer 3.5 Alpha 1.0 in WinXP, although the same thing happens
> on 3.4 running on a WinVista machine.
>
> Thanks,
>
> Justin Adams
>
> Department of Biomedical Sciences
> Grand Valley State University
> Allendale, MI 49401
> [hidden email]
> and
>
> School of Anatomical Sciences
> University of the Witwatersrand
> 7 York Road
> Parktown
> Johannesburg, RSA  2193
>
>
>
>
> ------------------------------
>
> Message: 7
> Date: Wed, 03 Feb 2010 22:31:05 -0500
> From: Luke Bloy <[hidden email]>
> Subject: [slicer-users] hide hidden folders in File dialogs
> To: [hidden email]
> Message-ID: <[hidden email]>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
> Hi all,
>
> This is probably silly, but often pressing "cntl h" in a file
> open
> dialog window toggles the display of hidden files (starting with
> a . on
> *nix systems) on and off but this has no effect in slicer3 (with
> KWwidgets).  Is there an option to hide hidden files and
> directories
> from the file dialogs in Slicer3?
>
> thanks,
> Luke
>
>
> ------------------------------
>
> _______________________________________________
> slicer-users mailing list
> [hidden email]
> http://massmail.spl.harvard.edu/mailman/admin/slicer-users
> To unsubscribe: send email to slicer-users-
> [hidden email] with unsubscribe as the subject
>
> End of slicer-users Digest, Vol 59, Issue 2
> *******************************************
_______________________________________________
slicer-users mailing list
[hidden email]
http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users
To unsubscribe: send email to [hidden email] with unsubscribe as the subject
Reply | Threaded
Open this post in threaded view
|

Re: slicer image registration - scaling problem

Dominik Meier
Hi Zane

if your scaling issue is due to incorrectly defined voxel sizes in the image headers, you can try adjust that in the "Volumes" module: simply enter the new correct voxel size where it says "Image Spacing". You should see the image stretch immediately. But be aware that this changes the physical representation of the image and will affect measurements etc.



If the scaling issue is due to registration, i.e. you would like to adjust the registration scaling factor, you can try to run an Affine registration between the two volumes that will take scale into account. Unfortunately the manual registration in the "Transforms" module does not (yet) allow to modify scale directly. You could do the above again, but that would scale along the original axis orientation, which would probably not useful. So you would have to apply your current registration (i.e. resample) and then change the scaling. Another option would be to save the transform and then edit.

I'd be happy to have a look at your data if you wish, if you can make an anonymized version.
We have a time-limited registration service and your case sound like a great example. If interested let me know. Details here:


Best,
Dominik

______________________________________________
DOMINIK MEIER, PH.D.
Assistant Professor of Radiology, Harvard Medical School
Center for Neurological Imaging, Brigham & Women's Hospital
221 Longwood Avenue, RF396, Boston, MA 02115
Phone: 617 278-0800
Fax: 617 264-5154




_______________________________________________
slicer-users mailing list
[hidden email]
http://massmail.spl.harvard.edu/mailman/listinfo/slicer-users
To unsubscribe: send email to [hidden email] with unsubscribe as the subject