[Insight-users] Re: ITK advice needed

Luis Ibanez luis.ibanez at kitware.com
Mon Aug 22 13:08:22 EDT 2005


Hi Siddarth,

Welcome to ITK,

You will find useful to follow the examples
in the ITK Software Guide:

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


You will find there typical values for the
parameters of most filters.


If an ITK program failed when reading your
data, chances are that you are providing
a file with an image format that is not
supported by ITK.  Please refer to the chapter
"Reading and Writing Images" for a full list
of the image file formats supported in ITK.


You will also find useful to read the material
of the ITK Tutorials:

    http://www.itk.org/HTML/Tutorials.htm


Please let us know if you have further questions,


   Thanks


     Luis



-------------------------
Siddharth Vikal wrote:
> Hello Luis,
>  
> I'm working on Dr. Gabor's project of Brachytherapy seed segmentation.
>  
> I've no previous experience with ITK, but I'm learning to get some. I'm 
> trying to use ITK with Visual Studio .Net. I've compiled the library in 
> a different directory using CMake as directed in the documentation. To 
> get familiar, I was trying to run the example programs. They build fine, 
> but for example, Derivative Image Filter example, when I give it an 
> input file (one from example data or my own, it doesn't matter), the 
> program gives a run time error & terminates.
>  
> I tried to trace & debug, the error seems to be coming at 
> filter->update() method during some memory allocation I guess...So I"m 
> unable to proceed further. This is true even for the 
> GradientMagnitudeFilter.
>  
> Also, in the documentation for DerivativeImageFilter, it is not clear 
> what values & what range of values is accepted in arguments of 
> SetOrder(int..) & SetDirection (int.) functions, I can just imagine that 
> I have to supply unsigned int values & nothing more.
>  
> Can you please clear my doubts.
>  
> Thanking you in anticipation and looking forward to hear from you.
>  
> regards
> siddharth
> 
> */Luis Ibanez <luis.ibanez at kitware.com>/* wrote:
> 
> 
>     Hi Gabor,
> 
> 
>     As a complement of what Stephen already described, here are some
>     names of filters and ITK examples that you may find useful to
>     look at for approaching your problem.
> 
> 
>     1) For detection of BBS
> 
>     You could use filters such as
> 
> 
>     http://www.itk.org/Insight/Doxygen/html/classitk_1_1HessianRecursiveGaussianImageFilter.html
>     and variations of the Hessian space like
>     http://www.itk.org/Insight/Doxygen/html/classitk_1_1Hessian3DToVesselnessMeasureImageFilter.html
> 
>     You could also use the Hat filters
> 
>     http://www.itk.org/Insight/Doxygen/html/classitk_1_1HMaximaImageFilter.html
> 
> 
> 
>     2) For calibration correction,
>     once you gather the images from grids that Stephen suggested,
>     you can use the following classes in order to compute
>     correction fields:
> 
>     http://www.itk.org/Insight/Doxygen/html/classitk_1_1KernelTransform.html
> 
>     and any of its derived classes:
>     http://www.itk.org/Insight/Doxygen/html/classitk_1_1ElasticBodyReciprocalSplineKernelTransform.html
>     http://www.itk.org/Insight/Doxygen/html/classitk_1_1ElasticBodySplineKernelTransform.html
>     http://www.itk.org/Insight/Doxygen/html/classitk_1_1ThinPlateR2LogRSplineKernelTransform.html
>     http://www.itk.org/Insight/Doxygen/html/classitk_1_1ThinPlateSplineKernelTransform.html
>     http://www.itk.org/Insight/Doxygen/html/classitk_1_1VolumeSplineKernelTransform.html
> 
> 
>     You could also use BSplines with the classes
>     http://www.itk.org/Insight/Doxygen/html/classitk_1_1BSplineDeformableTransform.html
> 
> 
> 
>     3) One application that you may want to look at is
> 
> 
>     InsightApplications/
>     IntensityBased2D3DRegistration
> 
> 
>     This code is available in the InsightApplications module
>     that you get with the releases, and that yo! u can get
>     through CVS too.
> 
> 
> 
>     You probably want to take a look also at
> 
> 
>     InsightApplications/
>     Curve2DExtractor
> 
> 
>     and
> 
>     InsightApplications/
>     Curve3DExtractor
> 
> 
> 
>     4) The 3D/2D registration problem for the treatment seeds is probably
>     better approached using the combination of the Levenberg-Marquardt
>     optimizer
>     http://www.itk.org/Insight/Doxygen/html/classitk_1_1LevenbergMarquardtOptimizer.html
> 
>     and the Rigid3DPerpective transform
>     http://www.itk.org/Insight/Doxygen/html/classitk_1_1Rigid3DPerspectiveTransform.html
> 
>     along with a Metric of your choice, that in the case of BBS, and
>     extracted features may be something like the
>     http://www.itk.org/Insight/Doxygen/html/classitk_1_1EuclideanDistancePointMetric.html
> 
> 
> 
> 
>     As Stephen already pointed out, there are many different ways
>     of approaching these problems, and there are many classes in ITK
>     that could be used for implementing those app! roaches. In a typical
>     application you probably will end up using many of the existing ITK
>     methods, and then customizing one or two for your particular type
>     of clinical problem.
> 
> 
> 
> 
> 
>     Please let us know if you have further questions,
> 
> 
> 
>     Regards,
> 
> 
> 
>     Luis
> 
> 
> 
>     -------------------------
>     Gabor Fichtinger wrote:
> 
>      > Dear Steve and Luis,
>      >
>      > Few days ago I talked with Bill Lorensen in Betheada about a clinical
>      > software problem that my group must solve rather urgently. (We
>     must go
>      > clinical during the fall, so we work on an accelerated time scale
>     here.)
>      > Bill suggested that I should seek your expert advice in the matter.
>      >
>      > The project involves reconstruction of the location of implanted
>     prostate
>      > brachytherapy seeds from a limited number of C-arm fluoroscopy
>     shots. The
>      > image processing work at hand entails segmentation of various
>     metal objects:
>      > fid! ucials (ellipses, straight line segments, BBs) and seeds
>     (5x1 mm metal
>      > pellets)
>      > from C-arm fluoroscopic images. Calibration fixtures and
>     dewarping grids are
>      > also need be segmented from the C-arm images. I have the following
>      > questions:
>      >
>      > (1) What filters and segmentation tools are available in ITK to
>     support
>      > these tasks?
>      >
>      > (2) Are there any fluoro calibration/distortion correction
>     modules in or
>      > outside ITK that we could use as an example/template?
>      >
>      > (3) Are there any application-level ITK examples that we could
>     use as an
>      > example/template?
>      >
>      > Steve, if I am not mistaken, you have done extensive work on
>     fluoro/CT
>      > registration of liver images that might have involved some of the
>     problems
>      > we are facing here. Please copy my associates on your reply. Our lead
>      > software engineer is Anton Deguet whom you probably remember from
>     various
>      > ITK work! shops.
>      >
>      > Thanks so very much, in advance, and have a great weekend.
>      >
>      > --gabor
>      >
>      > --
>      > Gabor Fichtinger, Ph.D.
>      > Associate Research Professor of
>      > Computer Science, Mechanical Engineering and Radiology
>      > Director of Engineering
>      > Center for Computer-Integrated Surgery
>      > Johns Hopkins University
>      > 3400 North Charles Street
>      > Baltimore MD 21218-2682
>      > New Engineering Bldg, Room B26
>      > Office: (410) 516-4057
>      > Mobile: (410) 562-6955
>      > Fax : (410) 516-5553
>      > mailto:gabor at cs.jhu.edu
>      > Assistant: Luwanna Spells
>      > (410) 516-6235, lspells at cs.jhu.edu
>      >
>      >
>      >
> 
> 
> 
> ------------------------------------------------------------------------
> Start your day with Yahoo! - make it your home page 
> <http://us.rd.yahoo.com/evt=34442/*http://www.yahoo.com/r/hs>



More information about the Insight-users mailing list