<div dir="ltr"><div>Hi Andreas,</div><div> </div><div>Thanks for the answer.</div><div> </div><div>The type of model I had in mind was an unstructured grid with somewhere between 10^4 and 10^5 (probably tetrahedral) cells.</div>
<div> </div><div>The idea is to define it as an elastic model, have an initial shape, and then during runtime to have a relatively small number (e.g. 100) boundary constraints that will induce a deformation.</div><div> </div>
<div>I am glad to hear of the Dune school. I will consider this idea.</div><div> </div><div>Regards</div><div> </div><div>Ofri Sadowsky<br></div><div class="gmail_quote">On Tue, Mar 20, 2012 at 12:42 AM, Dedner, Andreas <span dir="ltr"><<a href="mailto:A.S.Dedner@warwick.ac.uk">A.S.Dedner@warwick.ac.uk</a>></span> wrote:<br>
<blockquote style="margin:0px 0px 0px 0.8ex;padding-left:1ex;border-left-color:rgb(204,204,204);border-left-width:1px;border-left-style:solid" class="gmail_quote">
<div>
<div style="font-family:Tahoma;font-size:10pt;direction:ltr">Hi,<br>
<br>
thats really a rather difficult question to answer - even<br>
if I did know exactly what your model is. But without exactly the model and its size...<br>
<br>
I think the Galleries on the dune homepage give you a good<br>
idea of some of the things that have been done with dune -<br>
or at least of what is directly available.<br>
<br>
If dune is efficient enough - depends on the size of your<br>
problem and on how complex your model is to be. <br>
If your problem involves<br>
some type of computation using unstructured grids then dune should be similar in performance to other packages (as far as<br>
I know). If you can use Cartesian grids and finite-difference methods then probably dune is not the efficient choice.<br>
<br>
That's the best information I can give. There are some dune school which you might want to attend there you can discuss your problem directly with people developing dune. There is a scbool in Heidelberg, Germany (quite sone), there is one in Warwick, UK in
July and in Stuttgart sometime later this year.<br>
<br>
Best<br>
Andreas<br>
<div style="font-family:Times New Roman;font-size:16px">
<hr>
<div style="direction:ltr"><font color="#000000" face="Tahoma"><b>From:</b> dune-bounces+a.s.dedner=<a href="mailto:warwick.ac.uk@dune-project.org" target="_blank">warwick.ac.uk@dune-project.org</a> [dune-bounces+a.s.dedner=<a href="mailto:warwick.ac.uk@dune-project.org" target="_blank">warwick.ac.uk@dune-project.org</a>] on behalf of Ofri Sadowsky [<a href="mailto:sadowsky.o.phd@gmail.com" target="_blank">sadowsky.o.phd@gmail.com</a>]<br>
<b>Sent:</b> Monday, March 19, 2012 7:10 PM<br>
<b>To:</b> <a href="mailto:dune@dune-project.org" target="_blank">dune@dune-project.org</a>; <a href="mailto:dune-fem@dune-project.org" target="_blank">dune-fem@dune-project.org</a><br>
<b>Cc:</b> Moshe Shoham<br>
<b>Subject:</b> [Dune] Soft Tissue Deformation Models with Dune<br>
</font><br>
</div><div><div class="h5">
<div></div>
<div>
<div dir="ltr">
<div>To the Dune community,</div>
<div> </div>
<div>I am participating in a project involving tracking and estimating the motions and deformations of the brain in an open-skull surgery. Essentially, we have a stereo video stream, in which we detect and track feature points, and estimate their 3D locations.
These locations would be used first to define a rigid motion, and their residuals will define boundary conditions for the deformation.</div>
<div> </div>
<div>I would like to know if Dune contains the functionality needed for such an objective, and if there is a performance estimation of the computation on reasonable PC workstation (possibly multi-processor). Of course, there is a variety of ways to define
the model type and solution method. The simplest one we had in mind was a mesh of springs. But we can use more sophisticated models if they are useful.</div>
<div> </div>
<div>I will be happy to add more information on the problem if it is requested.</div>
<div> </div>
<div>Sincerely,<br clear="all">
<br>
-- <br>
</div>
<div dir="ltr">
<div>Ofri Sadowsky, PhD<br>
Computer Science Consulting and Training<br>
7 Carmel St., #37</div>
<div>Rehovot 76305</div>
<div>Israel</div>
<div> </div>
<div>Tel: <a href="tel:%2B972-77-3436003" target="_blank" value="+972773436003">+972-77-3436003</a><br>
Mob: <a href="tel:%2B972-54-3113572" target="_blank" value="+972543113572">+972-54-3113572</a></div>
</div>
<br>
</div>
</div>
</div></div></div>
</div>
</div>
</blockquote></div><br><br clear="all"><br>-- <br><div dir="ltr"><div>Ofri Sadowsky, PhD<br>Computer Science Consulting and Training<br>7 Carmel St., #37</div>
<div>Rehovot 76305</div>
<div>Israel</div>
<div> </div>
<div>Tel: +972-77-3436003<br>Mob: +972-54-3113572</div></div><br>
</div>