<div>Hi Kishore,</div>
<div> </div>
<div>Thanks for your answer.</div>
<div>I improved a lot the segmentation reducing the shape therm in the levelset equation<br>so that the weight of the image term is bigger than before, and now the results are more </div>
<div>precise then before so we are quite satisfied.</div>
<div> </div>
<div>Thank for your idea trying several orientation and then find the most common one, I will try to </div>
<div>see if the speed is acceptable for our application. I thought to solve this problem (the initial rotation)</div>
<div>by trying to correlate the average shape coming from the training set with the image.</div>
<div> </div>
<div>Anyway do you know of other works or papers that adds the shape prior to the levelsets or</div>
<div>can you point me to other shape based segmentation technique ?</div>
<div> </div>
<div>-Petin<br></div>
<div class="gmail_quote">On Fri, Sep 24, 2010 at 6:55 PM, Kishore Mosaliganti <span dir="ltr"><<a href="mailto:kishoreraom@gmail.com">kishoreraom@gmail.com</a>></span> wrote:<br>
<blockquote style="BORDER-LEFT: #ccc 1px solid; MARGIN: 0px 0px 0px 0.8ex; PADDING-LEFT: 1ex" class="gmail_quote">Hi Petin,<br><br>I have tried using the ShapePriorLevelSet filter before.<br><br>It is true that if you eliminate the registration aspect, then the<br>
final result will be better. This is because of fewer parameters to<br>estimate for finding the global optimal set. Giving a registration<br>initialization is a good idea where it is possible. You can also try<br>to automate that by specifying different starting orientations and<br>
choosing the most common orientation observed (via voting etc). That<br>may help reduce the registration burden on the user.<br><br>The convergence of the level-set is due to various terms including the<br>gradient. What you visually see as the object boundary from intensity<br>
may not be the location of the minima in the second gradient etc. The<br>levelset terms are attracted to a gradient basin. Therefore, you may<br>be perceiving the difference. The boundary is always going to be<br>difficult because of the partial voluming effects of neighboring<br>
voxels. I think your result is fairly good from my perspective.<br><br>Kishore<br>
<div>
<div></div>
<div class="h5"><br>On Thu, Sep 16, 2010 at 2:55 AM, petin <<a href="mailto:petin75@gmail.com">petin75@gmail.com</a>> wrote:<br>> Dear Itk users,<br>><br>> I'm using GeodesicActiveContourShapePriorLevelSetFilter image filter to<br>
> segment 2d sagittal<br>> images of vertebrae from L1 to S1. We created an average shape as described<br>> within the<br>> itk programming guide using about 50 images for each vertebrae, so the<br>> segmented vertebrae<br>
> were registered ,we did the pca and so on....<br>> After the training set has been created we tried the algorithm on some<br>> image, the user was<br>> able to click on the vertebrae and from this seed point the segmentation<br>
> started. We found that the<br>> final resul is a lot influenced by the rotation of the object to be<br>> segmented, than we modified a bit<br>> our user interface so that the user could specify an initial rotation of the<br>
> vertebrae to be segmented.<br>> The results we got were quite good, but still there are some cases like<br>> those attached where the<br>> final segmentation is not perfectly adherent to the vertebrae border and the<br>
> segmented object<br>> is not perfectly rotated.<br>><br>> Could you please give me some advice ?<br>> Any hint would help.<br>> Kind regards<br>> -Loris<br></div></div>> _____________________________________<br>
> Powered by <a href="http://www.kitware.com/" target="_blank">www.kitware.com</a><br>><br>> Visit other Kitware open-source projects at<br>> <a href="http://www.kitware.com/opensource/opensource.html" target="_blank">http://www.kitware.com/opensource/opensource.html</a><br>
><br>> Kitware offers ITK Training Courses, for more information visit:<br>> <a href="http://www.kitware.com/products/protraining.html" target="_blank">http://www.kitware.com/products/protraining.html</a><br>><br>
> Please keep messages on-topic and check the ITK FAQ at:<br>> <a href="http://www.itk.org/Wiki/ITK_FAQ" target="_blank">http://www.itk.org/Wiki/ITK_FAQ</a><br>><br>> Follow this link to subscribe/unsubscribe:<br>
> <a href="http://www.itk.org/mailman/listinfo/insight-users" target="_blank">http://www.itk.org/mailman/listinfo/insight-users</a><br>><br>><br></blockquote></div><br>