[Insight-users] Examples of...

[Luis Ibanez]

Hi Kendra,

Thanks for your clarifications,


If you already know the coordinates, then this is well-suited for using
any of the region growing algorithms that take multiple seeds:

                   ConnectedThreshold
                   ConfidenceConnected

It is also well suited for the watershed filter proposed in the
following paper in the Insight Journal:

http://insight-journal.org/InsightJournalManager/view_reviews.php?pubid=92

by Richard Beare and Gaetan Lehmann.



You may also need the connected components image filter
and the relabel image filter.

http://www.itk.org/Insight/Doxygen/html/classitk_1_1ConnectedComponentImageFilter.html
http://www.itk.org/Insight/Doxygen/html/classitk_1_1RelabelComponentImageFilter.html


    Regards,


        Luis


--------------
Kendra wrote:
> I don't need detect de nodules, I have the coordinates and type. I only 
> need create another image with this nodules marked with diferent colours.
> 
> 2008/3/28, Luis Ibanez <luis.ibanez at kitware.com 
> <mailto:luis.ibanez at kitware.com>>:
> 
> 
>     Hi Esther,
> 
>     You may not find all this functionalities in a single application.
>     However, you will find bits and pieces in different applications.
> 
>     1) You have it solved already, but just for the record,
>       one good example is
> 
>          InsightApplications/LiveTumorSegmentation
> 
>       and of course, you should consider Slicer (www.slicer.org
>     <http://www.slicer.org>).
> 
> 
>     2) By "mark" do you mean:
> 
>       a) having a user manually draw on the image ?
>       b) having an semi-automatic algorithm identify the nodules ?
> 
>       If you are talking about (a) you could look at SNAP
>       (http://www.itksnap.org/). Slicer (www.slicer.org
>     <http://www.slicer.org>) may
>       also be helpful here.
> 
>       If you are talking about (b)... this is still a research
>       area. Other people in the list may point you to publications
>       (hopefully Open Access ones...) describing suitable algorithms.
> 
> 
> 
>     3) Displaying an image in a Qt GUI is illustrated in the
> 
>         InsightApplications/Auxiliary/QtImageViewer
> 
>       You will also find useful to look at IGSTK (www.igstk.org
>     <http://www.igstk.org>)
>       which is based on ITK and VTK, and supports both FLTK and Qt
>       as GUI.
> 
>       Another option is to look at MITK (http://www.mitk.org/)
>       that also supports Qt, and is based on ITK and VTK as well.
> 
> 
> 
>      Regards,
> 
> 
>         Luis
> 
> 
>     ----------------
>     Kendra wrote:
> 
>         Please suggest me if there is any ITK (and VTK) application or
>         example that does something similar to this:
> 
>         1. Read a dicom serie of a pulmonar TAC. ( This is no problem,
>         I've got it).
>         2. Mark the nodules in RGB colors and labels with the type of
>         the nodule over the image too. (The location and type of the
>         nodules I've got in a XML file that I have parsed).
>         3. Display the result dicom serie in a Qt GUI.
> 
>         I've got an application that resolves the most of this questions
>         but probably  it's no the best way to do it and I like to see
>         better examples that gave me ideas for improve my applicaction
>         or correct it.
> 
>         Thanks a lot.
> 
>         -- 
>         Esther.
> 
> 
>         ------------------------------------------------------------------------
> 
>         _______________________________________________
>         Insight-users mailing list
>         Insight-users at itk.org <mailto:Insight-users at itk.org>
>         http://www.itk.org/mailman/listinfo/insight-users
> 
> 
> 
> 
> -- 
> Esther.