<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=Content-Type content="text/html; charset=iso-8859-1">
<META content="MSHTML 6.00.2900.2802" name=GENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY
style="WORD-WRAP: break-word; khtml-nbsp-mode: space; khtml-line-break: after-white-space"
bgColor=#ffffff>
<DIV><FONT face=Arial size=2>Hi Nick,</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT> </DIV>
<DIV><FONT face=Arial size=2>thanks a lot for your help. </FONT><FONT face=Arial
size=2>I will try it with your values again.</FONT></DIV>
<DIV><FONT face=Arial size=2>Can you perhaps explain the parameters alpha and
gamma</FONT></DIV>
<DIV><FONT face=Arial size=2>a little bit more detailed for me, since I didn't
find information about the </FONT><FONT face=Arial size=2>real meaning of these
parameters.</FONT></DIV>
<DIV><FONT face=Arial size=2>Nearly all the posts only deal with elasticity and
rho.</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT> </DIV>
<DIV><FONT face=Arial size=2>Another question is:</FONT></DIV>
<DIV><FONT face=Arial size=2>If I change <FONT face="Times New Roman"
size=3>elasticity to 1.0e4 and Rho to 1.0e4, than the deformations
should</FONT></FONT></DIV>
<DIV>even become bigger (I used 1.0e5 and the
deformation is already much to high).</DIV>
<DIV><FONT face=Arial size=2></FONT> </DIV>
<DIV><FONT face=Arial size=2>Also one has to specify rho and rhoc in the filter,
but as far as I remember, rhoC should always</FONT></DIV>
<DIV><FONT face=Arial size=2>be approx. 1.0 and equals elasticity divided
by rho (or vice versa), so why is it specified separately?</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT> </DIV>
<DIV><FONT face=Arial size=2>Best regards,</FONT></DIV>
<DIV><FONT face=Arial size=2>Markus</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT> </DIV>
<DIV><FONT face=Arial size=2></FONT> </DIV>
<BLOCKQUOTE
style="PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV style="FONT: 10pt arial">----- Original Message ----- </DIV>
<DIV
style="BACKGROUND: #e4e4e4; FONT: 10pt arial; font-color: black"><B>From:</B>
<A title=ntustison@gmail.com href="mailto:ntustison@gmail.com">Nicholas
Tustison</A> </DIV>
<DIV style="FONT: 10pt arial"><B>To:</B> <A title=m.weigert@fz-juelich.de
href="mailto:m.weigert@fz-juelich.de">Markus Weigert</A> </DIV>
<DIV style="FONT: 10pt arial"><B>Cc:</B> <A title=insight-users@itk.org
href="mailto:insight-users@itk.org">insight-users@itk.org</A> </DIV>
<DIV style="FONT: 10pt arial"><B>Sent:</B> Thursday, April 06, 2006 3:28
PM</DIV>
<DIV style="FONT: 10pt arial"><B>Subject:</B> Re: [Insight-users] Setting up
FEM - Registration parameters</DIV>
<DIV><FONT face=Arial size=2></FONT><FONT face=Arial size=2></FONT><FONT
face=Arial size=2></FONT><FONT face=Arial size=2></FONT><FONT face=Arial
size=2></FONT><FONT face=Arial size=2></FONT><BR></DIV>
<DIV>Hi Markus,</DIV>
<DIV><FONT face=Arial size=2></FONT><FONT face=Arial size=2></FONT><FONT
face=Arial size=2></FONT><FONT face=Arial size=2></FONT><FONT face=Arial
size=2></FONT><FONT face=Arial size=2></FONT><BR
class=khtml-block-placeholder></DIV>
<DIV>The registration scenario you describe seems reasonably straightforward.
However, there are a couple of items I noticed from the images that you
might want to change. Generally, you want to keep alpha = 0.5 (as
opposed to 1.0) so that the solution is stable for the time step that you
choose. The chosen material governs the regularization of your
registration results. The default values that I have are Elasticity =
1.0e4 and Rho = 1.0e4 which give me reasonable registration results. You
should be able to achieve reasonable results using the specified elastic
material. Unfortunately, without running the registration myself, it is
difficult to know exactly what the problem is. </DIV>
<DIV><FONT face=Arial size=2></FONT><FONT face=Arial size=2></FONT><FONT
face=Arial size=2></FONT><FONT face=Arial size=2></FONT><FONT face=Arial
size=2></FONT><FONT face=Arial size=2></FONT><FONT face=Arial
size=2></FONT><FONT face=Arial size=2></FONT><BR
class=khtml-block-placeholder></DIV>
<DIV>On a related note, it is interesting that you chose to forgo the .config
file. I actually rewrote the itkFEMRegistrationFilter and plan to
include it in an upcoming release along with additional modifications and
extensions to the library (including B-spline elements). The first thing
I got rid of was the .config file along with some parameters that weren't
being used.</DIV>
<DIV><FONT face=Arial size=2></FONT><FONT face=Arial size=2></FONT><FONT
face=Arial size=2></FONT><FONT face=Arial size=2></FONT><FONT face=Arial
size=2></FONT><FONT face=Arial size=2></FONT><FONT face=Arial
size=2></FONT><FONT face=Arial size=2></FONT><BR
class=khtml-block-placeholder></DIV>
<DIV>Nick Tustison </DIV>
<DIV><FONT face=Arial size=2></FONT><FONT face=Arial size=2></FONT><FONT
face=Arial size=2></FONT><FONT face=Arial size=2></FONT><FONT face=Arial
size=2></FONT><FONT face=Arial size=2></FONT><FONT face=Arial
size=2></FONT><BR class=khtml-block-placeholder></DIV>
<DIV><FONT face=Arial size=2></FONT><FONT face=Arial size=2></FONT><FONT
face=Arial size=2></FONT><FONT face=Arial size=2></FONT><FONT face=Arial
size=2></FONT><FONT face=Arial size=2></FONT><BR
class=khtml-block-placeholder></DIV><BR>
<DIV>
<DIV>On Apr 6, 2006, at 7:36 AM, Markus Weigert wrote:</DIV><BR
class=Apple-interchange-newline>
<BLOCKQUOTE type="cite">Hi,<BR><BR>after scanning the list on FEM -
registration, I still have huge problems <BR>to configure the paramters for
my specific registration problem with the FEM - method.<BR><BR>I want to
register two 3D MR 256^3 Thorax images with only small deformation.<BR>The
problem is, that however I set the parameters <B>elasticity</B> and <B>rho /
rhoc</B>,<BR>it doesn't deliver any useful results.<BR><BR>I started with
elasticity 100000 (even higher than in the examples), but the image became
completely<BR>deformed. Also, it spends only 2 iterations in the process
(has this something to do with<BR>the energyreductionfactor???) and after
the registration, the metric (I use SSD) is much bader then before.<BR>I
think this should not be the case, after what I understood about the theory
behind this method.<BR>Also with elasticity 1000.000, it still gets
completely deformed.<BR>With even higher the deformation decreases of
course, but the missregistration of the lungs is never<BR>caught (see middle
of the images at the bottom of this message).<BR><BR>Please note, that I
don't use a config file but set all the parameters manually.<BR>Also I
wonder what's the meaning of the parameters alpha and gamma (changing
these doesn't<BR>seem to have a great effect on the result).<BR><BR>Indeed,
I call the following methods to setup the FEMRegistrationFilter (public
functions),<BR>while I didn't find a function to set the scaling, which is
also used in the config file (is this intended???): <BR><FONT
color=#ff0000><BR>
m_registrationFilter->SetFixedImage(m_fixedImage);<BR>
m_registrationFilter->SetMovingImage(m_floatingImage);<BR><BR>
m_registrationFilter->DoMultiRes(true); <BR><BR>
m_registrationFilter->SetMeshPixelsPerElementAtEachResolution(this->m_meshPixPerElement);<BR><BR>
m_registrationFilter->SetNumberOfIntegrationPoints(this->m_numberOfIntegrationPoints);<BR><BR>
m_registrationFilter->SetRho(this->m_rho);<BR><BR>
m_registrationFilter->SetElasticity(this->m_elasticity);<BR>
<BR> m_registrationFilter->SetGamma(this->m_gamma);<BR>
<BR>
m_registrationFilter->SetWidthOfMetricRegion(this->m_widthOfMetricRegion);<BR>
m_registrationFilter->ChooseMetric((int) this->m_metric);<BR>
m_registrationFilter->SetMaximumIterations(this->m_maxNumberOfIterations,0);<BR><BR>
m_registrationFilter->SetTimeStep(this->m_timestep);<BR>
m_registrationFilter->SetLineSearchMaximumIterations(this->m_maxLineSearchIterations);<BR>
m_registrationFilter->SetEnergyReductionFactor(this->m_energyReductionFactor);<BR>
<BR> if(m_metric==3)
m_registrationFilter->SetDescentDirectionMaximize();<BR> else{
m_registrationFilter->SetDescentDirectionMinimize();}<BR>
<BR> m_registrationFilter->SetAlpha(this->m_alpha);<BR>
m_registrationFilter->DoLineSearch(2);<BR>
m_registrationFilter->EmployRegridding(1);<BR>
m_registrationFilter->UseLandmarks(false);<BR>
m_registrationFilter->SetWriteDisplacements(false);</FONT><BR><BR>Also
what is about the material properties:<BR>Does it make sense to change
something here??? <BR><BR> <FONT
color=#ff0000>itk::fem::MaterialLinearElasticity::Pointer m;<BR> m =
itk::fem::MaterialLinearElasticity::New();<BR> m->GN =
0;
// Global number of the material<BR> m->E =
m_registrationFilter->GetElasticity(); // Young's modulus -- used
in the membrane<BR> m->A =
1.0;
// Cross-sectional area<BR> m->h =
1.0;
// Thickness<BR> m->I =
1.0;
// Moment of inertia<BR> m->nu =
0.;
// Poisson's ratio -- DONT CHOOSE 1.0!!<BR> m->RhoC =
this->m_density; // Density<BR> <BR> // Create the
element type <BR> Element3DType::Pointer
e1=Element3DType::New();<BR>
e1->m_mat=dynamic_cast<itk::fem::MaterialLinearElasticity*>( m
);<BR> m_registrationFilter->SetElement(e1);<BR>
m_registrationFilter->SetMaterial(m);</FONT><BR><BR><BR>Additionally, you
can find some screenshots which show the problem and the
RegistrationFilter<BR>outputs here:<BR><BR><A class=moz-txt-link-freetext
href="http://www.markusweigert.de/images/E10%5E8_FEMRegOutput_iter2.jpg">http://www.markusweigert.de/images/E10%5E8_FEMRegOutput_iter2.jpg</A><BR><A
class=moz-txt-link-freetext
href="http://www.markusweigert.de/images/E10%5E8_FEMRegOutput_iter1.jpg">http://www.markusweigert.de/images/E10%5E8_FEMRegOutput_iter1.jpg</A><BR><A
class=moz-txt-link-freetext
href="http://www.markusweigert.de/images/before_registration.jpg">http://www.markusweigert.de/images/before_registration.jpg</A><BR><A
class=moz-txt-link-freetext
href="http://www.markusweigert.de/images/E10%5E8_FEMRegOutput_iter2.jpg">http://www.markusweigert.de/images/E10%5E8_FEMRegOutput_iter2.jpg</A><BR><A
class=moz-txt-link-freetext
href="http://www.markusweigert.de/images/result_E100000.jpg">http://www.markusweigert.de/images/result_E100000.jpg</A><BR><A
class=moz-txt-link-freetext
href="http://www.markusweigert.de/images/result_E1000000.jpg">http://www.markusweigert.de/images/result_E1000000.jpg</A><BR><A
class=moz-txt-link-freetext
href="http://www.markusweigert.de/images/result_E10000000.jpg">http://www.markusweigert.de/images/result_E10000000.jpg</A><BR><BR><BR>I
hope somebody will have the time to answer at least some of my
questions.<BR><BR><BR>Best regards,<BR>Markus<BR><BR><BR><BR><BR><BR><BR>
<DIV
style="MARGIN: 0px">_______________________________________________</DIV>
<DIV style="MARGIN: 0px">Insight-users mailing list</DIV>
<DIV style="MARGIN: 0px"><A
href="mailto:Insight-users@itk.org">Insight-users@itk.org</A></DIV>
<DIV style="MARGIN: 0px"><A
href="http://www.itk.org/mailman/listinfo/insight-users">http://www.itk.org/mailman/listinfo/insight-users</A></DIV></BLOCKQUOTE></DIV><BR></BLOCKQUOTE></BODY></HTML>