[Insight-users] Error on running Diffeomorphic Demons method

Tom Vercauteren tom.vercauteren at gmail.com
Tue Mar 23 05:01:34 EDT 2010


Dear Ganesh,

The error message says it all I guess. You need more RAM in order to
run the algorithm on your data.

You can try to use
  --update-rule=2
which should consume a little less memory. However, it is only a
marginal change. My current implementation is not really focused on
keeping a low memory footprint. This could be a nice project though.

Regarding your other questions, "MSE" is the similarity criterion. It
is simply the sum of squared differences between the registered images
divided by the number of voxels that have been used in the
computation. "harmo" is a the harmonic energy, a measure of the
non-smoothness of the deformation field. I invite you to check out the
paper for more information.

Hope this helps,
Tom


On Mon, Mar 22, 2010 at 20:48, Ganesh Narayanasamy
<nganesh76 at hotmail.com> wrote:
>
> http://www.insight-journal.org/browse/publication/154
> While registering a pair of 3D-CTs using Diffeomorphic Demons method and
> most of the default parameters, I noticed an error that cropped up about
> 2/3rd the way through. This is a 3 level resolution registration process and
> the error cropped up after the 2nd level saying "Failed to allocate memory
> for image.". Can you guide me to complete the entire registration using this
> method. The error message displayed on the screen is shown below (briefly).
>
> Can someone also educate me what is meant here by MSE (mean square error?)
> and also the "harmo".
>
>
> Thanks,
> Ganesh
>
> ************************************************************************
> "What the mind of man can conceive and believe, It can achieve" - Napolean
> Hill
> -Ganesh Narayanasamy,Ph.D., PostDoc-University of Kentucky Cancer Center
> Earlier work: http://www.sciencedaily.com/releases/2007/05/070504133017.htm
> ************************************************************************
>
> ERROR MESSAGE:
> 1:
> 2:
> .
> .
> .
> 14: MSE 9.60677e+008 - harmo. 0.00562125 - max|Jac| 1.53167 - min|Jac|
> 0.659633
> - ratio(|Jac|<=0) 0
> Finished Multi-resolution iteration :0
> ==============================
>
> 1:
> 2:
> .
> .
> 9: MSE 9.59992e+008 - harmo. 0.00805274 - max|Jac| 1.75971 - min|Jac|
> 0.639913 -
>  ratio(|Jac|<=0) 0
>
> Finished Multi-resolution iteration :1
> ==============================
>
> Unexpected error.
>
> itk::ExceptionObject (0178F5DC)
> Location: "class itk::Vector<float,3> *__thiscall
> itk::ImportImageContainer<unsi
> gned long,class itk::Vector<float,3> >::AllocateElements(unsigned long)
> const"
> File: c:\ganesh\itk316\itk\code\common\itkImportImageContainer.txx
> Line: 188
> Description: Failed to allocate memory for image.
>
>
>
>
> ________________________________
> Hotmail: Trusted email with Microsoft’s powerful SPAM protection. Sign up
> now.
> _____________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Kitware offers ITK Training Courses, for more information visit:
> http://www.kitware.com/products/protraining.html
>
> Please keep messages on-topic and check the ITK FAQ at:
> http://www.itk.org/Wiki/ITK_FAQ
>
> Follow this link to subscribe/unsubscribe:
> http://www.itk.org/mailman/listinfo/insight-users
>
>


More information about the Insight-users mailing list