[Insight-users] Re: reg similarity measure(deformable registration)

[Luis Ibanez]

Hi Dalal,

You can use Mutual Information in ITK in the
following two Deformable Registration methods:

   - FEM-Based deformable registration

   - BSplines deformable transform


--


Demons *does not* support Mutual Information since
from its very nature, it is an optical-flow based method,
therefore expecting to have image of the same modality and
with the same intensity distribution.

Note that there is in fact a large family of algorithms
that belong to the Demons category. The two implementations
  that are available in ITK are based on the paper by
J.P. Thirion entitled "Fast Non-Rigid Matching of 3D Medical
Images", May 1995.


--


The example for BSplines was committed in CVS on april 19th 2004.
http://www.itk.org/cgi-bin/viewcvs.cgi/Examples/Registration/DeformableRegistration4.cxx?root=Insight

Therefore you will not see this example in the ITK
source tree unless you use a recent CVS version of
ITK. You will certainly not find it in ITK 1.6.

The example was stabilized by May 22 2004, so you
will be safer if you download a CVS version posterior
to that date.


Note that ITK 1.8 is scheduled to be released in
August.


The difference between the code in the Testing directory
and the code in the Examples directory is that in Testing
most (if not all) of the class functionalities are exercised.
The test are not supposed to execute any useful functionality
in themselves, just make sure that the methods of the class
work as advertised.

The example, on the other hand, are didactic pieces of
code with abundant comments that show the simplest use
of the classes. Most examples can actually be used as command
line tools for solving real image processing tasks.  This is
the case for all the registration examples in

         Insight/Examples/Registration/

Whenever you have the option of learning from an example
and from a test, you should choose the example since it
was conceived as a teaching aid. Most testing programs
will actually look obscure to the newcomers.



The only example that requires parameter files is the
FEM-Based deformable registration. The parameter files
are in the directory


     Insight/Examples/Data/


That is explained in detail in the Chapter on image
registration (that you have not read yet) in the
SoftwareGuide

      http://www.itk.org/ItkSoftwareGuide.pdf

Chapter 8, Section 8.12, pdf-page 318.







    Regards,


       Luis



--------------------
Dalal, Dhaval wrote:

>  
> thanx luis
> i will be going through this and will get back to you.
> but i need to know
> - are there any mutual information based deformabale registration avaiable
> in itk?
> in examples/registration 
> - deformable registration 1 - fem 
>   deformable registration 2 -  demons
> i dont see where bslines is ?
> - wat is the difference between the code in the test folder and the example
> folder?
> - where are the parameterfiles listed?
> can u help me there.
> regards
> dhaval
> -----Original Message-----
> From: Luis Ibanez
> To: Dalal, Dhaval
> Cc: Insight-users at itk.org
> Sent: 7/11/2004 2:14 PM
> Subject: Re: reg similarity measure(deformable registration)
> 
> 
> 
> Hi Dalal,
> 
> Please read the Tutorials in Image Registration available at
> 
>        http://www.itk.org/HTML/Tutorials.htm
> 
> in particular
> 
> http://www.itk.org/CourseWare/Training/RegistrationMethodsOverview.pdf
> http://www.itk.org/CourseWare/Training/NonRigidRegistrationMethods.pdf
> 
> You should also read the chapter on Image Registration in
> the ITK SoftwareGuide
> 
>           http://www.itk.org/ItkSoftwareGuide.pdf
> 
> --
> 
> 
> About your questions:
> 
> 
> 1) Yes,
> 
>     You can use deformable registration without landmarks.
>     As a matter of fact, none of the deformable registration
>     methods that you listed require landmarks to be defined.
> 
>     You can optionally use landmarks in FEM if you feel
>     inclined to do so.
> 
> 
> 2) "What method should give better results" ?
> 
>     That's a retoric question that only make sense in the
>     "paper-publishing" context of "my method is better than
>     yours".   That's only useful in the Matrix...
> 
>     In real life there is no such a thing as a "better method",
>     since the performance of each method is extremly dependent
>     on the parameters that you select for it. What you will find
>     in real life is the ubiquitous trade-off of quality versus cost.
> 
>     Where cost involves computation time, memory allocation, user-time
>     required for fine tunning parameters, and of course, amount of
>     liters of coffee to be consumed in the process.
> 
> 
>     So, turn the question around and ask yourself first:
> 
> 
>      - How many weeks do I have to get this registration done ?
> 
>      - How long should a registration computation take ?
> 
>      - How many weekends and sleepless nights am I planning
>        to pass fine tunning the parameters of the registration ?
> 
>      - How many journal and conference papers am I willing to
>        study in order to get the background needed for getting
>        the best out of this method ?
> 
>      - How much registration precision is required for the
>        application for which this registration is intended ?
> 
> 
>     Your choice of Deformable Registration method will be
>     quite straight forward once you answer those questions   :-)
> 
> 
>     Here are some guidelines
>     that you may want to consider at that point:
> 
>     Demons will be appropriate if the deformations in the
>     image are small. Where "small" is defined as "being in
>     the range of the transition regions of the anatomical
>     structures in the image". That is, if the transition in
>     intensity between one tissue and another in your image
>     is 3 pixels wide,... then  Demons will be limited to
>     correct for deformation equal or smaller than 3 pixels.
>     You can of course overcome this limitation by using
>     Demons in a Multi-Resolution context.
> 
>     There are few parameters in Demons, but the
>     computational load is high.
> 
> 
> 
>     BSplines have also few parameters to fit, mostly those of
>     the optimizer.
> 
>     FEM is the most flexible method, probably the one from
>     which you can get better quality results, but also the
>     one that will require more effort on your part to get
>     it working correctly in your images.
> 
> 
>     I would suggest you to start with BSplines, using a small
>     grid (e.g. 8x8x8), and if needed move to a larger grid.
>     Using a larger grid will be needed only if your images
>     have local deformations that go in different directions.
> 
> 
> 
> 
> 3) You are right in the first statement and wrong in
>     the second.
> 
>     - Yes,
>       Mutual-Information have prove to be appropriate for
>       multi-modality registration.
> 
>     - Yes,
>       ITK *does* provide implementations of the MutualInformation
>       metric.
> 
>       There are actually four different implementation
>       of Mutual Information in ITK
> 
>          - Viola Wells
> http://www.itk.org/Insight/Doxygen/html/classitk_1_1MutualInformationIma
> geToImageMetric.html
> 
>          - Mattes et al.
> http://www.itk.org/Insight/Doxygen/html/classitk_1_1MattesMutualInformat
> ionImageToImageMetric.html
> 
>          - KullbackLeibler
> http://www.itk.org/Insight/Doxygen/html/classitk_1_1KullbackLeiblerCompa
> reHistogramImageToImageMetric.html#_details
> 
>          - Histogram based
> http://www.itk.org/Insight/Doxygen/html/classitk_1_1MutualInformationHis
> togramImageToImageMetric.html
> 
> 
> 
> 4) You will find the full list of image metrics available
>     in ITK in the Doxygen documentation.
> 
>  
> http://www.itk.org/Insight/Doxygen/html/group__RegistrationMetrics.html
> 
>     As you can see there are twelve different Image-to-Image
>     Metrics available in ITK at this point.
> 
>     You are certainly welcome to write new implementations of
>     Image metrics. If you decide to do so, we strongly encourage
>     you to contribute your code back to ITK so other users can
>     take advantage of your efforts.
> 
> 
> 
> 
> Regards,
> 
> 
>     Luis
> 
> 
> 
> --------------------
> Dalal, Dhaval wrote:
> 
>> 
>>hey luis
>>I know that ITK provides multiple methods for performing deformable
>>registration. 
>>
>>In particular: 
>>
>>1) BSpline Transform 
>>2) Demons deformable registration 
>>3) FEM-Based deformable registration 
>>
>>I want to register 2 MRI datasets, their shapes are quite different 
>>hence i will use deformabale registration method.
>>my questions are
>>1) I do not have any fixed landmarks, is it possible to register
> 
> without
> 
>>   known landmarks in these 3 methods. 
>>2) which of these 3 methods should give the better result.
>>3) Regarding the similarity metric, i think it is an accepted fact
> 
> that 
> 
>>   mutual information gives the better result for multi-modal
> 
> registration. 
> 
>>itk i do not think provides a framework for mutual information based
>>registration? why is that so.
>>4) other similarity measures assume the intensity to be the same in
> 
> both the
> 
>>images. not mutual information.
>>what are the similarity measures implemented in ITK.
>>is it possible to write the code for mutual information.
>>Kindly help me out please.
>>your help will be greatly appreciated.
>>regards
>>Dhaval Dalal 
>>
> 
> 
> 
>