Hi Lingfeng,
from what I see, the tubes I think you're referring to (but correct me if I'm
wrong) are related to
external metal structures and they should be very bright.
If this is the case, I'd suggest you threshold the image (you have to do it
outside vmtk, there's no
script for thresholding in vmtk for some odd reason :-) ), like in 3DSlicer
(www.slicer.org))
and then run the vessel enhancement filter. Even though thresholding would
preserve the outer
edges of the external tubes, essentially leaving hollow tubes, the enhancement
filter should wipe
them out.
Hope this helps
Luca
On Sep 28, 2012, at 4:11 PM, Lingfeng Zhuang wrote:
> Dear Luca,
>
> Thank you for your very detailed explaination.
>
> I tried tune gamma. The result is much better now. After doing vessel
> enhancement, the vessel is clear in MIP viewer. I have a further question.
>
> Applying MarchingCube filter with lower iso value(0.1) procedures all
> vessels. But many of vessels are connected with tube composed of background
> noise. So, even after applying vtkPolyDataConnectivityFilter to PolyData,
> some tubes composed of background noise will still be kept in final
> segmentation result.
> Applying MarchingCube filter with higher iso value(0.3) will earse most thin
> vessels.
>
> What should I do to keep most thin vessels, in the mean time, dropping all
> tubes composed of background noise?
> <MIP.png>
> <iso_0.1.png>
> <iso_0.3.png>
>
> Best regards,
>
> zlf
>
> On Mon, Sep 24, 2012 at 9:45 PM, Luca Antiga <luca.ant...@orobix.com> wrote:
> Hi Lingfeng,
> teeth are very bright and tube-like, which is why they get enhanced.
>
> One (kind of hacky) thing you could do is to remap the intensity above the
> level
> of the brightest vessels, so that very bright objects become very dark ones,
> and the vessel enhancement filters will not detect them (they could detect
> the background to object transitions, but whether this might create an actual
> issue
> will have to be evaluated in practice).
>
> One more thing you can do is to try to be a lot more selective with the
> sigmas
> (the spatial scales). Since the filter is scale-specific and the scale is
> expressed
> (in your case) in mm, what you're doing with your command line is that you are
> "erasing" what is smaller than 1 mm. I expect most intracranial vessels to be
> below
> that level, so maybe you could fiddle with the sigmamin and see what happens.
>
> One last aspect is about the other parameters, alpha, beta and gamma. In
> particular,
> try varying gamma, which modulates what the filter sees as background noise vs
> structure to enhance. You might want to tune that one so that vessels, which
> are
> a lot dimmer than bones, are enhanced as well. Try varying it by orders of
> magnitude.
>
> Keep in mind that you're working with an unmasked CT dataset. This means that
> a few vessels (I'm thinking about the cavernous segment of the ICA) will be
> really
> close to bones (or in this case surrounded by them), so the filter will not
> be able
> to enhance them accurately (since their actual scale will not be detected
> correctly
> as they form a vessel+bone complex in intensity space).
>
> Hope this helps
>
> Luca
>
>
> --
> Luca Antiga, PhD
> Cofounder and Principal Scientist, Orobix Srl
> via L.A. Muratori 3, 24123 Bergamo, Italy
>
> orobix: www.orobix.com
> home: lantiga.github.com
> twitter: twitter.com/lantiga
>
> On Sep 19, 2012, at 5:03 PM, Lingfeng Zhuang wrote:
>
>> Hi all,
>>
>> I tried use vmtkimagevesselenhancement to enhancement vessel. But I found no
>> vessel is enhanced, instead, bone (like tooth) is enhanced.
>>
>> The script run is
>>
>> vmtkimagecast -ifile "Brain_CTA.mhd" -type float --pipe
>> vmtkimagevesselenhancement -method sato -sigmamax 2 -sigmasteps 2 --pipe
>> vmtkimagemipviewer
>>
>> Regards,
>>
>> zlf
>>
>> ------------------------------------------------------------------------
>> Live Security Virtual Conference
>> Exclusive live event will cover all the ways today's security and
>> threat landscape has changed and how IT managers can respond. Discussions
>> will include endpoint security, mobile security and the latest in malware
>> threats.
>> http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/_______________________________________________
>> vmtk-users mailing list
>> vmtk-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/vmtk-users
>
>
> ------------------------------------------------------------------------------
> Got visibility?
> Most devs has no idea what their production app looks like.
> Find out how fast your code is with AppDynamics Lite.
> http://ad.doubleclick.net/clk;262219671;13503038;y?
> http://info.appdynamics.com/FreeJavaPerformanceDownload.html_______________________________________________
> vmtk-users mailing list
> vmtk-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/vmtk-users
------------------------------------------------------------------------------
Don't let slow site performance ruin your business. Deploy New Relic APM
Deploy New Relic app performance management and know exactly
what is happening inside your Ruby, Python, PHP, Java, and .NET app
Try New Relic at no cost today and get our sweet Data Nerd shirt too!
http://p.sf.net/sfu/newrelic-dev2dev
_______________________________________________
vmtk-users mailing list
vmtk-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/vmtk-users
