<html><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">hi eleni,<div><br></div><div>you can monitor cvs changes in the review directory using this link :</div><div><a href="http://public.kitware.com/cgi-bin/viewcvs.cgi/Code/Review/?root=Insight&sortby=date#dirlist">http://public.kitware.com/cgi-bin/viewcvs.cgi/Code/Review/?root=Insight&sortby=date#dirlist</a></div><div><br></div><div>There should be a lot of changes in the next days, so it might be unpractical for us to let you know everytime we commit a change. We will try to do it though.</div><div><br></div><div>cheers.</div><div><br></div><div>alex.</div><div><br></div><div><br></div><div><br><div><div>On Jun 7, 2009, at 11:28 PM, Sgouritsa, Eleni wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite"> <div> <!-- Converted from text/plain format --><p><font size="2">Thank you very much.<br> Yes, I think too that my problem was coming from the foreground term.<br> Is it possible to notify me when you finish with the other changes of the class so I can work directly with the corrected version?<br> <br> Regards,<br> Eleni<br> <br> <br> -----Original Message-----<br> From: Arnaud Gelas [<a href="mailto:arnaud_gelas@hms.harvard.edu">mailto:arnaud_gelas@hms.harvard.edu</a>]<br> Sent: Sun 6/7/2009 6:48 PM<br> To: Sgouritsa, Eleni<br> Cc: Kishore Mosaliganti; <a href="mailto:insight-users@itk.org">insight-users@itk.org</a>; Luis Ibanez<br> Subject: Re: [Insight-users] itkScalarChanAndVeseSparseLevelSetImageFilter<br> <br> Dear Eleni,<br> <br> I have corrected a bug in the formula of the foreground constant <br> calculation. This error could really affect the results of your <br> segmentation process.<br> You can get the new version from the cvs repository.<br> <br> Note that this code may change a lot in the next couple of days <br> (Kishore and I will be fixing various parts of the code).<br> <br> Arnaud<br> <br> On Jun 7, 2009, at 6:12 PM, Sgouritsa, Eleni wrote:<br> <br> > What I understand is that the result should be independent of how we <br> > define the distance function (negative inside or outside). And if we <br> > initialize the level set totally inside the data , the level set <br> > should expand according to equation (7) of the paper where you <br> > describe the method, whereas if we initialize outside it should <br> > shrink. In insight journal it happens like that but not in the itk <br> > release.<br> > Thank you very much,<br> > Eleni<br> ><br> ><br> > -----Original Message-----<br> > From: Sgouritsa, Eleni<br> > Sent: Sun 6/7/2009 3:30 PM<br> > To: Kishore Mosaliganti<br> > Cc: Luis Ibanez; <a href="mailto:insight-users@itk.org">insight-users@itk.org</a><br> > Subject: RE: [Insight-users] <br> > itkScalarChanAndVeseSparseLevelSetImageFilter<br> ><br> > Hello,<br> > thank you very much for the help. Unfortunately I work with medical <br> > data that I am not supposed to give.<br> > Can you explain a little more point 2. You mean that I have to <br> > define my level set as inside positive and outside negative to have <br> > the opposite propagation direction?<br> > I tried it but it did not work. I both cases, if I initialize the LS <br> > inside the boundary of my data, it shrinks. I use mu=0, nu=0, l1=1 <br> > and l2=1. Do I need negative l1?When I initialize almost the whole <br> > LS outside my data it seems to work in some cases.Generally, I get <br> > different results from the insight journal that I was getting correct.<br> ><br> > Thank you,<br> > Eleni<br> ><br> ><br> > -----Original Message-----<br> > From: Kishore Mosaliganti [<a href="mailto:kishoreraom@gmail.com">mailto:kishoreraom@gmail.com</a>]<br> > Sent: Fri 6/5/2009 1:02 PM<br> > To: Sgouritsa, Eleni<br> > Cc: Luis Ibanez; <a href="mailto:insight-users@itk.org">insight-users@itk.org</a><br> > Subject: Re: [Insight-users] <br> > itkScalarChanAndVeseSparseLevelSetImageFilter<br> ><br> > Hi Eleni,<br> ><br> > Thank you for pointing out these problems. You are right that the <br> > Review<br> > code is not very stable as yet. We are working on fixing them. <br> > During the<br> > transfer from the IJ paper to the Review directory, a lot of <br> > modifications<br> > were made. We might have inadvertently introduced new errors.<br> ><br> > Regarding your problems:<br> ><br> > 2. The level-set function is defined by convention as inside is <br> > negative and<br> > outside is positive. So the negative values become your foreground and<br> > positive values become the background. This controls the propagation<br> > direction.<br> ><br> > 3. I am looking at the Halt function. Looks like the boundary case <br> > of 0<br> > iterations is not coded right.<br> ><br> > 4. Regarding the high RMS values, the level-set function is re-<br> > initialized<br> > to a signed distance function after every few iterations. It so <br> > happens that<br> > the default setting is 1 iteration. Hence, the RMS change is being<br> > calculated after the reinitialization while it needs to be done <br> > before. We<br> > are also working on correcting this. Thank you for pointing this out.<br> ><br> ><br> > Is it possible for you to post a small sample of your data for us to <br> > work on<br> > the errors.<br> ><br> ><br> > Kishore<br> ><br> > On Tue, Jun 2, 2009 at 11:24 AM, Sgouritsa, Eleni<br> > <<a href="mailto:esgourit@central.uh.edu">esgourit@central.uh.edu</a>>wrote:<br> ><br> > > Hello,<br> > > I m sorry if my point was not clear. The problems that I am facing <br> > using<br> > > this class to segment 3D data are the following:<br> > ><br> > > 1. My data has negative origin (for example (-100, -100, -60). <br> > Unless I do<br> > > SetOrigin(0,0,0) (or I guess any origin >=0) my program crashes. I <br> > have to<br> > > set my negative origin as zero for my program to run. Does the <br> > class assumes<br> > > positive origin of the input image?<br> > ><br> > > 2. I initialize my level set inside (so the evolution should be <br> > outwards).<br> > > In this case I can not make my level set expand (it can just <br> > shrink, or not<br> > > move). I think I should use something like <br> > SetReverseExpansionDirection to<br> > > control the propagation direction, but I can not find this option <br> > in the<br> > > class public members. (This is currently my main problem where I <br> > am stuck,<br> > > any help would be very useful!)<br> > ><br> > > 3. The maximum RMS Error serves (along with the number of <br> > iterations) as a<br> > > stopping criteria inside the Halt function (of<br> > > itkMultiphaseFiniteDifferenceImageFilter.txx file).<br> > > Lets say I set maximum RMS Error = 0.01.<br> > > The Halt function returns the following:<br> > > return ( (this->GetElapsedIterations() >= this-<br> > >m_NumberOfIterations) ||<br> > > ( this->GetMaximumRMSError() >= m_RMSChange ) );<br> > > While the Halt function returns false the evolution continues.<br> > > However, in iteration 0, I get m_RMSChange = 0, so the evolution <br> > always<br> > > stops doing 0 iterations (since the condition this-<br> > >GetMaximumRMSError() >=<br> > > m_RMSChange becomes true)<br> > > If I modify the Halt function to returm the following:<br> > > if(this->GetElapsedIterations()==0)<br> > > return false;<br> > > else<br> > > return ( (this->GetElapsedIterations() >=<br> > > this->m_NumberOfIterations) || ( this->GetMaximumRMSError() >= <br> > m_RMSChange )<br> > > );<br> > > then it works.<br> > > The other problem is that RMSChange in the next iterations <br> > (1,2,3...) has<br> > > very big values (like 5000) even near convergence. So the only way <br> > to stop<br> > > the evolution is through the iterations number.<br> > ><br> > > I am still investigating the class and I am not absolutely sure <br> > about the<br> > > points mentioned but I would like to share with you for more advice.<br> > ><br> > ><br> > > Thank you very much,<br> > > Lena<br> > ><br> > ><br> > ><br> > ><br> > > -----Original Message-----<br> > > From: Luis Ibanez [<a href="mailto:luis.ibanez@kitware.com">mailto:luis.ibanez@kitware.com</a><<a href="mailto:luis.ibanez@kitware.com">luis.ibanez@kitware.com</a><br> > ><br> > > ]<br> > > Sent: Mon 6/1/2009 8:39 PM<br> > > To: Sgouritsa, Eleni<br> > > Cc: alex gouaillard; <a href="mailto:insight-users@itk.org">insight-users@itk.org</a><br> > > Subject: Re: [Insight-users] <br> > itkScalarChanAndVeseSparseLevelSetImageFilter<br> > ><br> > ><br> > > Hi Lena,<br> > ><br> > > Could you please be more specific about your problem report on<br> > > item (2) ?<br> > ><br> > > Is there a chance that you can report this in a form that<br> > > we could reproduce ?<br> > ><br> > ><br> > > Thanks<br> > ><br> > ><br> > > Luis<br> > ><br> > ><br> > > -----------------------------<br> > > Sgouritsa, Eleni wrote:<br> > > > Hi,<br> > > ><br> > > > I have two more questions.<br> > > > 1. In the insight journal code there was an option<br> > > > SetReverseExpansionDirection to control if the expansion of the <br> > level<br> > > > set will be inwards or outwards. How can I do it using the<br> > > > itkScalarChanAndVeseSparseLevelSetImageFilter of 3.14 release?<br> > > ><br> > > > 2. I still have a problem when my 3D data have negative origin.<br> > > ><br> > > > Thanks and regards,<br> > > > Lena<br> > > ><br> > > > -----Original Message-----<br> > > > From: Sgouritsa, Eleni<br> > > > Sent: Mon 6/1/2009 4:00 AM<br> > > > To: alex gouaillard<br> > > > Cc: <a href="mailto:insight-users@itk.org">insight-users@itk.org</a><br> > > > Subject: RE: [Insight-users]<br> > > itkScalarChanAndVeseSparseLevelSetImageFilter<br> > > ><br> > > > Hello,<br> > > > finally the second issue was my fault (I just had a problem while<br> > > > including the header files of the review folder).<br> > > > For the first, the inconsistency is just a matter of name. The <br> > code runs<br> > > > correctly, just the name assignement in the txx is a little <br> > misleading.<br> > > ><br> > > > Thank you very much,<br> > > > Lena<br> > > ><br> > > ><br> > > > -----Original Message-----<br> > > > From: alex gouaillard <br> > [<a href="mailto:alexandre_gouaillard@hms.harvard.edu">mailto:alexandre_gouaillard@hms.harvard.edu</a><<a href="mailto:alexandre_gouaillard@hms.harvard.edu">alexandre_gouaillard@hms.harvard.edu</a><br> > ><br> > > ]<br> > > > Sent: Sun 5/31/2009 11:34 PM<br> > > > To: Sgouritsa, Eleni<br> > > > Cc: <a href="mailto:insight-users@itk.org">insight-users@itk.org</a><br> > > > Subject: Re: [Insight-users]<br> > > itkScalarChanAndVeseSparseLevelSetImageFilter<br> > > ><br> > > > hi eleni,<br> > > ><br> > > > we'll take a look tomorrow.<br> > > ><br> > > > thanks for the feedback.<br> > > ><br> > > > alex.<br> > > ><br> > > > On May 31, 2009, at 11:52 PM, Sgouritsa, Eleni wrote:<br> > > ><br> > > > > Hello,<br> > > > > I know that the review folder of the new release (3.14) is not<br> > > > > stable yet but I would just like to give two comments:<br> > > > ><br> > > > > 1. I found an inconsistency between the template arguments of <br> > the<br> > > > > itkScalarChanAndVeseSparseLevelSetImageFilter.h and the<br> > > > > itkScalarChanAndVeseSparseLevelSetImageFilter.txx.<br> > > > > Specifically, in the header file, the template is :<br> > > > > template < class TInputImage, class TFeatureImage, class<br> > > > > TOutputImage, class TFunction,<br> > > > > class TSharedData, typename TIdCell = unsigned int ><br> > > > > whereas in the txx it is:<br> > > > > template < class TInput, class TFeature, class TFunction,<br> > > > > class TOutputPixel, class TSharedData, typename TIdCell ><br> > > > > There is a change in the order of the 3rd and 4th argument.<br> > > > ><br> > > > > 2. I can not find this class<br> > > > > (itkScalarChanAndVeseSparseLevelSetImageFilter) incuded in the<br> > > > > CMakeLists of the review folder<br> > > > ><br> > > > > Regards,<br> > > > > Lena<br> > > > ><br> > > > > <ATT00001.txt><br> > > ><br> > > ><br> > > ><br> > > ><br> > > > <br> > ------------------------------------------------------------------------<br> > > ><br> > > > _____________________________________<br> > > > Powered by <a href="http://www.kitware.com">www.kitware.com</a><br> > > ><br> > > > Visit other Kitware open-source projects at<br> > > > <a href="http://www.kitware.com/opensource/opensource.html">http://www.kitware.com/opensource/opensource.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">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">http://www.itk.org/mailman/listinfo/insight-users</a><br> > ><br> > ><br> > > _____________________________________<br> > > Powered by <a href="http://www.kitware.com">www.kitware.com</a><br> > ><br> > > Visit other Kitware open-source projects at<br> > > <a href="http://www.kitware.com/opensource/opensource.html">http://www.kitware.com/opensource/opensource.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">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">http://www.itk.org/mailman/listinfo/insight-users</a><br> > ><br> > ><br> ><br> ><br> ><br> > <ATT00001.txt><br> <br> <br> </font> </p> </div> <span><ATT00001.txt></span></blockquote></div><br></div></body></html>