[Insight-users] Geodesic Active Contours Segmentation with Shape Guidance

[Zachary Pincus]

Hello Steffi,

I again strongly encourage reading the Leventon et al. paper in which  
this method is described (available here: http:// 
splweb.bwh.harvard.edu:8000/pages/papers/leventon/cvpr00/cvpr00.pdf )  
before attempting to use the ITK filters for the same. The paper is  
short and clear, and the filters have a lot of complexity that needs  
to be understood before they can be used properly.

The "Mean shape" is not a binary segmented object: it is the mean of  
a population of segmented objects, all represented as signed distance  
maps. Similarly, the mode shapes are PCA-derived "shape modes" of an  
entire population of shapes. ITK has a class to calculate these  
images from a population: http://www.itk.org/Doxygen/html/ 
classitk_1_1ImagePCAShapeModelEstimator.html . I have always thought  
that it was something of a shape that this stuff isn't spelled out  
more clearly in the documentation for that example, but it is a  
pretty long example as-is.

Zach Pincus

On Feb 6, 2006, at 3:43 AM, Steffi Meyer wrote:

> Hi everyone,
>
> I have a question about the geodesic active contours. In 2D,
>
> meanShapeReader->SetFileName( argv[13] );
>
> this argv[13] is a filename (e.g. geodesic.tiff), which contains a
> segmented object e.g. liver (object white, background black). And
>
> fileNamesCreator->SetSeriesFormat( argv[15] );
>
> which SeriesFormat? At the moment I use fileNamesCreator- 
> >SetSeriesFormat("E:\\shape%d.tiff").
> Is this right? And in the loop of the original file I got a runtime  
> error "Runtime error! This application has requested the Runtime to  
> terminate it in an unusual way."
>
> for ( unsigned int k = 0; k < numberOfPCAModes; k++ )
>     {
>     ReaderType::Pointer shapeModeReader = ReaderType::New();
>     shapeModeReader->SetFileName( shapeModeFileNames[k].c_str() );
>     shapeModeReader->Update();
>     shapeModeImages[k] = shapeModeReader->GetOutput();
>     }
>
> So I don't know exactly where the error is and if the assumptions  
> with the .tiff files are correct.
> Perhaps someone can help me with this. At the moment I want to try  
> the 2D version and then extend it to 3D.
>
>
> ______________________________________________________________________ 
> ____
> Erweitern Sie FreeMail zu einem noch leistungsstarkeren E-Mail- 
> Postfach!		
> Mehr Infos unter http://freemail.web.de/home/landingpad/?mc=021131
>
> _______________________________________________
> Insight-users mailing list
> Insight-users at itk.org
> http://www.itk.org/mailman/listinfo/insight-users