<div dir="ltr">Hi Zoro<br><div class="gmail_extra"><br><br><div class="gmail_quote">On Mon, Jan 20, 2014 at 3:03 AM, zoro <span dir="ltr"><<a href="mailto:k09tranhoan@gmail.com" target="_blank">k09tranhoan@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">Dear all,<br>
<br>
I am performing registration between 3D volumes (chest before and after<br>
breathing) using Mattes MI & LBFGSBOptimizer. The result is pretty good but<br>
I need to improve it. Range of values in fixed image and moving image is<br>
(0,4000) with about 50 histogram bins number.<br>
<br>
The purpose of my work is to register cardiac parts but when using MI, it<br>
also registers bone parts. I want to define the range of value in volume to<br>
perform Mattes MI. For example, only in (0,1000) without modified the<br>
original volume. By the way, I used mask to concentrate at cardiac part and<br>
the result is also good.<br>
<br>
I wonder there is a method or not to define (min,max) using in MI. I<br>
attached part of MattesMutualInformation in ITK as I thought it related to<br>
this problem.<br>
<br></blockquote><div><br></div><div> </div><div><br></div><div>It looks like you may find useful the following methods:</div><div><br></div><div>SetFixedImageSamplesIntensityThreshold()<br></div><div>UseFixedImageSamplesIntensityThresholdOn()<br>
</div><div><br></div><div><a href="http://www.itk.org/Doxygen/html/classitk_1_1ImageToImageMetric.html#ae18d98221d211736fc0448296e2c077c">http://www.itk.org/Doxygen/html/classitk_1_1ImageToImageMetric.html#ae18d98221d211736fc0448296e2c077c</a><br>
</div><div><a href="http://www.itk.org/Doxygen/html/classitk_1_1ImageToImageMetric.html#ae56734d7cc4bb45a6ba462e56f260546">http://www.itk.org/Doxygen/html/classitk_1_1ImageToImageMetric.html#ae56734d7cc4bb45a6ba462e56f260546</a><br>
</div><div><br></div><div><br></div><div>They allow you to define a minimum threshold to exclude pixels with </div><div>intensities lower than that, from participating in the computation of the</div><div>metric.</div><div>
<br></div><div>You could modify this code, in order to manage a range of intensities,</div><div>instead of just the lower threshold.</div><div><br></div><div>Places to look for inspiration are:</div><div><br></div><div><a href="https://github.com/InsightSoftwareConsortium/ITK/blob/master/Modules/Registration/Common/include/itkImageToImageMetric.hxx#L581">https://github.com/InsightSoftwareConsortium/ITK/blob/master/Modules/Registration/Common/include/itkImageToImageMetric.hxx#L581</a><br>
</div><div><br></div><div>and</div><div><br></div><div><a href="https://github.com/InsightSoftwareConsortium/ITK/blob/master/Modules/Registration/Common/include/itkImageToImageMetric.hxx#L637">https://github.com/InsightSoftwareConsortium/ITK/blob/master/Modules/Registration/Common/include/itkImageToImageMetric.hxx#L637</a><br>
</div><div><br></div><div><br></div><div><br></div><div> Regards,</div><div><br></div><div><br></div><div> Luis</div><div><br></div><div><br></div><div><br></div><div> </div></div></div></div>