[Insight-users] Supervised Benchmark : Comparing Segmentations

Tobias Heimann t.heimann at dkfz-heidelberg.de
Fri Jul 27 11:06:02 EDT 2007


Hi Grub,

You can also use the ITK-based sources we are using for our MICCAI 
segmentation workshop, see 
http://mbi.dkfz-heidelberg.de/grand-challenge2007/sites/down.htm

Best regards,
Tobias


Luis Ibanez wrote:

>
> Hi Grub,
>
> Please look at the following filters:
>
>
>    itkHausdorffDistanceImageFilter
>    itkSimilarityIndexImageFilter
>    itkContourMeanDistanceImageFilter
>
> and the metrics:
>
>    itkKappaStatisticImageToImageMetric
>    itkMatchCardinalityImageToImageMetric
>
>
>
> You may also find interesting to look at
> the code in:
>
>
>     InsightApplications/
>           TumorSegmentationEvaluation
>
>
>
>  Regards,
>
>
>     Luis
>
>
>
> -----------------
> grub boot wrote:
>
>>
>>     Note also that ITK provides several filters for
>>     generating metrics that compare two segmentations.
>>
>>
>>     Except for the STAPLE method, these comparision filters
>>     will not tell which segmentation is right, they simply
>>     compute "distances" between segmentations).
>>
>>
>> Actually I need those data sets to compare them with the segmentation 
>> results from
>> my algorithms. So I need some way to compare two segmentations of the 
>> same image.
>>
>> Can you please tell me where to look for the relevant filters and how 
>> to use them?
>>
>> Thanks in advance;
>
> _______________________________________________
> Insight-users mailing list
> Insight-users at itk.org
> http://www.itk.org/mailman/listinfo/insight-users
>

-- 
------------------------------------------------------------------------
Tobias Heimann                                 Email:  t.heimann at dkfz.de
Medical and Biological Informatics (E130)      URL:   www.dkfz.de/en/mbi
German Cancer Research Center
Im Neuenheimer Feld 280                        Phone: +49 (0)6221-423548
D-69120 Heidelberg, Germany                    Fax:   +49 (0)6221-422345
------------------------------------------------------------------------



More information about the Insight-users mailing list