<div dir="ltr">Hi Pierre,<div><br></div><div>We have been experimenting with 3D printer lately, </div><div>and will be happy to help.</div><div><br></div><div> <a href="http://www.kitware.com/blog/home/post/591">http://www.kitware.com/blog/home/post/591</a><br>
</div><div> <a href="http://www.kitware.com/blog/home/post/610">http://www.kitware.com/blog/home/post/610</a><br></div><div><br></div><div><br></div><div>One of the easiest methods to perform bone segmentation </div>
<div>is to use the region growing filters, in particular the:</div><div><br></div><div> ConnectedThresholdImageFilter<br></div><div><a href="http://www.itk.org/Doxygen/html/itkConnectedThresholdImageFilter_8h.html">http://www.itk.org/Doxygen/html/itkConnectedThresholdImageFilter_8h.html</a><br>
</div><div><br></div><div>Please keep in mind that segmentation requires a bit of</div><div>parameter tuning.</div><div><br></div><div>In the case of bone structures in CT, you typically can</div><div>expect the bones to have intensities in the range of</div>
<div><br></div><div> 600~700 HU</div><div><br></div><div>up to</div><div><br></div><div> 2000+ HU</div><div><br></div><div>This implies that you can approach the bone segmentation</div>
<div>by first placing a set of seeds on the brightest part of the </div><div>bone that you want to segment.</div><div><br></div><div>Set the Upper threshold to a very high value (3000, for example)</div><div>since normally there wouldn't be any brighter values in the image</div>
<div>than bone, (unless there are implant artifacts).</div><div><br></div><div>Set the Lower threshold to 800, and then run the filter.</div><div><br></div><div>If the resulting segmentation goes too far, then increase the value</div>
<div>of the lower threshold and rerun the filter.</div><div><br></div><div>Overall, the process is capturing bone pixels by going down</div><div>the levels of intensity in HU.</div><div><br></div><div> </div><div>There are many other segmentation techniques that would be </div>
<div>applicable here, but region growing thresholding is a good</div><div>place to start.</div><div><br></div><div><br></div><div> Please let us know if we can help with more details,</div><div><br></div><div><br></div>
<div> Thanks</div><div><br></div><div><br></div><div> Luis</div><div><br></div><div><br></div></div><div class="gmail_extra"><br><br><div class="gmail_quote">On Sun, Jan 19, 2014 at 9:08 AM, Pierre Starkov <span dir="ltr"><<a href="mailto:starkovpierre@yahoo.com" target="_blank">starkovpierre@yahoo.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div><div style="font-size:12pt;font-family:HelveticaNeue,Helvetica Neue,Helvetica,Arial,Lucida Grande,sans-serif">Hello,<br>
<br>I'm a med school student from Geneva, Switzerland, and i'm currently working on a plugin project for a software called Osirix (<a href="http://www.osirix-viewer.com/" target="_blank">http://www.osirix-viewer.com/</a>).<br>
I'm totally new to ITK, so i have a few questions.<br><br><div>My plugin has to segment bones from a set of ct-scan images, make it 3D, and send the whole to a 3D printer.</div><div>I wonder if you would have any suggestions of how to perform a nice and quick bone segmentation using ITK.</div>
<div>I mean, what would be the right functions to use, from the ITK-library for the segmentation part? :)</div><div><br></div><div style="font-style:normal;font-size:16px;background-color:transparent;font-family:HelveticaNeue,Helvetica Neue,Helvetica,Arial,Lucida Grande,sans-serif">
Hope you can help me in
my task.</div><div style="font-style:normal;font-size:16px;background-color:transparent;font-family:HelveticaNeue,Helvetica Neue,Helvetica,Arial,Lucida Grande,sans-serif">Best Regards,</div><div style="font-style:normal;font-size:16px;background-color:transparent;font-family:HelveticaNeue,Helvetica Neue,Helvetica,Arial,Lucida Grande,sans-serif">
<br></div><div style="font-style:normal;font-size:16px;background-color:transparent;font-family:HelveticaNeue,Helvetica Neue,Helvetica,Arial,Lucida Grande,sans-serif">Pierre Starkov<br></div></div></div><br>_______________________________________________<br>
Community mailing list<br>
<a href="mailto:Community@itk.org">Community@itk.org</a><br>
<a href="http://public.kitware.com/cgi-bin/mailman/listinfo/community" target="_blank">http://public.kitware.com/cgi-bin/mailman/listinfo/community</a><br>
<br></blockquote></div><br></div>