[vtkusers] vtkImageReslice class

[David Gobbi]

Hi Emmanouil,

One thing you can do is create a vtkImageImport object for each DICOM 
image, this will allow you to create a vtkImageData from each DICOM that 
you have stored in memory.  Then, you can use vtkImageAppend to create 
an image volume out of the series of images.  The output of 
vtkImageAppend can be used by vtkImageReslice.

 - David


Emmanouil Moschidis wrote:
> Before being able to use some parts of the code I have to solve a 
> different problem. (I posted a new question but either it was not 
> quite clear or nobody has done that before (?)  )
>
> I work within a DICOM viewer so the DICOM series are already loaded (I 
> do not read from a file like in case of the vtkImageReader2). Actually 
> every DICOM image is stored in a custom class and all together in a 
> NSArray (yes it is objective-C). So I want to load the images from the 
> NSArray to the vtkImageReader2 object... I am not sure how should I do 
> that.
>
> Maybe first I need to load them into a vtkImageData object? (still I 
> haven't found any example for that)
>
> And then again I have the problem how to get the output properly. But 
> I guess these questions do not belong to this class.
>
> */David Gobbi <dgobbi at atamai.com>/* wrote:
>
>     Hi Emmanouil,
>
>     The best example for what you are trying to do is probably this one:
>
>     VTK/Examples/ImageProcessing/Cxx/ImageSlicing.cxx
>
>     This example is only in the CVS version of VTK, it isn't in the VTK 5
>     release.
>
>     What I do with tracking systems is use the coordinate/quaternion from
>     the tracking system to create a vtkTransform, then I concatenate that
>     transform with a vtkMatrix4x4 from the patient registration, and then
>     feed the resulting transform to vtkImageReslice as the
>     ResliceTransform.
>
>     If you have looked at the class page for vtkImageReslice, you will
>     notice that there are two methods for setting orientations:
>     SetResliceAxes() and SetResliceTransform().
>
>     For tracking, I always set ResliceTransform to the
>     position/orientation
>     of the tool (in patient coordinates), and then set the ResliceAxes
>     according to how I want to orient the slice relative to
>     orientation of
>     the tool. In that way, the ResliceAxes is used to choose the tool
>     axes
>     along which you will be extracting the slice.
>
>     The tricky thing with vtkImageReslice is getting the OutputOrigin,
>     OutputSpacing, and OutputExtent set properly for your application.
>     Together, these parameters can be used to describe an output slice
>     plane
>     with a particular pixel spacing. So for example:
>
>     OutputOrigin = (-127.5, -127.5, 0.0)
>     OutputSpacing = (1.0, 1.0, 1.0)
>     OutputExtent = (0, 255, 0, 255, 0, 0)
>
>     These describe a 256x256 output slice with 1mm pixel spacing. The
>     first
>     pixel is at (-127.5,-127.5,0) in the "output coordinate system" of
>     vtkImageReslice. The centre of the slice is at (0,0,0) in the "output
>     coordinate system" of vtkImageReslice.
>
>     Now, the original input volume is in the "input coordinate system" of
>     vtkImageReslice. We can call the input coordinate system "x" and the
>     output coordinate system "x'". The relationship between these
>     coordinates is as follows:
>
>     x = T*M*x' where "T" is ResliceTransform and "M" is ResliceMatrix
>
>     So this tells you that, if "M" is chosen to be a pure rotation
>     matrix,
>     and if "T" contains the position of the tool point in input image
>     coordinates, then the point x' = (0,0,0), which is in the centre
>     of our
>     output plane, will correspond to the point x = tool_point. So, we
>     will
>     be extracting a slice through the tool point position.
>
>     I know that is probably not the best description, let me know if
>     things
>     aren't quite clear.
>
>     - David
>
>
>
>
>     Emmanouil Moschidis wrote:
>     > Hello
>     >
>     > I want to use the vtkImageReslice class for reading a series of
>     slices
>     > and exctracting one slice based on coordinates and quaternions
>     > received from a tracking system.
>     >
>     > Is there any specific documentation and/or any working examples for
>     > this class? It is very powerful yet the only usable information
>     I have
>     > found is the class reference webpage.
>     >
>     > Any hint about how to proceed towards the insight of it would be
>     > highly appreciated
>     >
>     > thanks
>     >
>     >
>     ------------------------------------------------------------------------
>     > Sick sense of humor? Visit Yahoo! TV's Comedy with an Edge
>     > to
>     > see what's on, when.
>     >
>     ------------------------------------------------------------------------
>     >
>     > _______________________________________________
>     > This is the private VTK discussion list.
>     > Please keep messages on-topic. Check the FAQ at:
>     http://www.vtk.org/Wiki/VTK_FAQ
>     > Follow this link to subscribe/unsubscribe:
>     > http://www.vtk.org/mailman/listinfo/vtkusers
>     >
>
>
> ------------------------------------------------------------------------
> Be a PS3 game guru.
> Get your game face on with the latest PS3 news and previews at Yahoo! 
> Games. <http://us.rd.yahoo.com/evt=49936/*http://videogames.yahoo.com>