[Elastix] Running multi-metric registration

Jihun Kim jihun at umich.edu
Sat Apr 6 00:46:18 CEST 2013


Thanks Marius,
Reducing (SP_a) value allowed registrations to work properly.

Regards,
Jihun
--
*Jihun Kim*
Ph. D. candidate
Mechanical Engineering
University of Michigan, Ann Arbor
734-276-7167


On Fri, Apr 5, 2013 at 3:10 AM, <M.Staring at lumc.nl> wrote:

>  Hi Jihun,****
>
> ** **
>
> Try reducing your step size. My rule of thumb is that the stepsize
> multiplied by the metric gradient should be in the order of 1.****
>
> ** **
>
> Marius****
>
> ** **
>
> *From:* elastix-bounces at bigr.nl [mailto:elastix-bounces at bigr.nl] *On
> Behalf Of *Jihun Kim
> *Sent:* donderdag 4 april 2013 21:35
> *To:* elastix at bigr.nl
> *Subject:* Re: [Elastix] Running multi-metric registration****
>
> ** **
>
> Sorry that I did not attach this file. ****
>
> ** **
>
> Regards,****
>
> Jihun
> ****
>
> --****
>
> *Jihun Kim*****
>
> Ph. D. candidate****
>
> Mechanical Engineering****
>
> University of Michigan, Ann Arbor****
>
> 734-276-7167****
>
> ** **
>
> On Thu, Apr 4, 2013 at 3:27 PM, Jihun Kim <jihun at umich.edu> wrote:****
>
> Dear Elastix-subscribers, ****
>
> ** **
>
> I am trying to run multi-metric registration with AdvancedMeanSquares and
> TransformRigidityPenalty. I attached a parameter file to run a
> registration. I have got the following a command window message: ****
>
> ** **
>
> =====Message
> Begins======================================================================================
> ****
>
> elastix is started at Thu Apr 04 15:20:05 2013. ****
>
>  ****
>
> elastix runs at: Jihun-PC ****
>
>   Windows 7, Service Pack 1 (Build 7601) ****
>
>   with 3956 MB memory, and 4 cores @ 2393 MHz. ****
>
> -------------------------------------------------------------------------
> ****
>
>  ****
>
> Running elastix with parameter file 0:
> "D:\elastix\hncbm8\ct7\staring-ms\Parameters.txt". ****
>
>  ****
>
> Current time: Thu Apr 04 15:20:05 2013. ****
>
> Reading the elastix parameters from file ... ****
>
>  ****
>
> Installing all components. ****
>
> InstallingComponents was successful. ****
>
>  ****
>
> ELASTIX version: 4.600 ****
>
> Command line options from ElastixBase: ****
>
> -f        D:\hncbm8\fect1-7.mhd ****
>
> -m        D:\hncbm8\ct7.mhd ****
>
> -fMask    unspecified, so no fixed mask used ****
>
> -mMask    unspecified, so no moving mask used ****
>
> -out      D:\elastix\hncbm8\ct7\staring-ms/ ****
>
> -p        D:\elastix\hncbm8\ct7\staring-ms\Parameters.txt ****
>
> -priority unspecified, so NORMAL process priority ****
>
> -threads  unspecified, so all available threads are used ****
>
> Command line options from TransformBase: ****
>
> -t0       D:\hncbm8\ct7toct1.txt ****
>
> WARNING: The parameter "BSplineTransformSplineOrder", requested at entry
> number 0, does not exist at all. ****
>
>   The default value "3" is used instead. ****
>
> WARNING: The parameter "UseCyclicTransform", requested at entry number 0,
> does not exist at all. ****
>
>   The default value "false" is used instead. ****
>
>  ****
>
> Reading images... ****
>
> Reading images took 489 ms. ****
>
>  ****
>
> Reading the elastix parameters from file ... ****
>
>  ****
>
> Initialization of all components (before registration) took: 43 ms. ****
>
> Preparation of the image pyramids took: 2882 ms. ****
>
>  ****
>
> Resolution: 0 ****
>
> WARNING: The parameter "ShowExactMetricValue", requested at entry number
> 0, does not exist at all. ****
>
>   The default value "false" is used instead. ****
>
> WARNING: The parameter "CheckNumberOfSamples", requested at entry number
> 0, does not exist at all. ****
>
>   The default value "true" is used instead. ****
>
> WARNING: The parameter "ShowExactMetricValue", requested at entry number
> 0, does not exist at all. ****
>
>   The default value "false" is used instead. ****
>
> WARNING: The parameter "CheckNumberOfSamples", requested at entry number
> 0, does not exist at all. ****
>
>   The default value "true" is used instead. ****
>
> Setting the fixed masks took: 0 ms. ****
>
> Setting the moving masks took: 0 ms. ****
>
> WARNING: The parameter "UseRelativeWeights", requested at entry number 0,
> does not exist at all. ****
>
>   The default value "false" is used instead. ****
>
> WARNING: The parameter "NumberOfSamplesForSelfHessian", requested at entry
> number 0, does not exist at all. ****
>
>   The default value "100000" is used instead. ****
>
> WARNING: The parameter "SelfHessianSmoothingSigma", requested at entry
> number 0, does not exist at all. ****
>
>   The default value "1" is used instead. ****
>
> WARNING: The parameter "SelfHessianNoiseRange", requested at entry number
> 0, does not exist at all. ****
>
>   The default value "1" is used instead. ****
>
> WARNING: The parameter "MaximumNumberOfSamplingAttempts", requested at
> entry number 0, does not exist at all. ****
>
>   The default value "0" is used instead. ****
>
> Elastix initialization of all components (for this resolution) took: 2 ms.
> ****
>
> Initialization of AdvancedMeanSquares metric took: 0 ms. ****
>
> Initialization of TransformRigidityPenalty term took: 0 ms. ****
>
> 1:ItNr 2:Metric 2:Metric0 2:Metric1 3:StepSize 4:||Gradient0||
> 4:||Gradient1|| 4:||Gradient|| Metric-LC Metric-OC Metric-PC Time0[ms]
> Time1[ms] Time[ms] ||Gradient-LC|| ||Gradient-OC|| ||Gradient-PC|| ****
>
> SGD ****
>
> elxTemplate ****
>
> 0 13224.394736 13224.394736 0.000000 7457.268480 593.852296 0.000000
> 593.852296 0.0000000000 0.0000000000 0.0000000000 103 115 295 0.0000000000
> 0.0000000000 0.0000000000 ****
>
> Time spent in resolution 0 (ITK initialisation and iterating): 0.333 s. **
> **
>
> Stopping condition: Error in metric. ****
>
>  ****
>
> itk::ExceptionObject (0178FA88) ****
>
> Location: "ElastixTemplate - Run()"  ****
>
> File:
> d:\src4\elastix_sources_v4.6\src\common\costfunctions\itkAdvancedImageToImageMetric.hxx
> ****
>
> Line: 628 ****
>
> Description: itk::ERROR: AdvancedMeanSquaresMetric(03ABCAB0): Too many
> samples map outside moving image buffer: 0 / 5000 ****
>
>  ****
>
> Error occurred during actual registration. ****
>
> ** **
>
> Errors occurred! ****
>
> =====Message
> Ends======================================================================================
> ****
>
> ** **
>
> Following the instruction given in FAQ, I applied an initial transform but
> I don't know what to try further. I am wondering if you could give me some
> suggestions. ****
>
> Thanks very much. ****
>
> ** **
>
> Regards,****
>
> Jihun****
>
> --****
>
> *Jihun Kim*****
>
> Ph. D. candidate****
>
> Mechanical Engineering****
>
> University of Michigan, Ann Arbor****
>
> 734-276-7167****
>
> ** **
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.bigr.nl/pipermail/elastix/attachments/20130405/f889fc53/attachment.html>


More information about the Elastix mailing list