--- Original Message ---
From : "Dana Brooks"<brooks@ece.neu.edu>
To : "Jess Tate"<jess@sci.utah.edu>
Cc : "Donghyeon Kim"<danielkim@gist.ac.kr>, scirun-users@sci.utah.edu
Date : 2011/10/17 월요일 오전 1:22:19
Subject : Re: [SCIRUN-USERS] Re: set electrode conductivity on DefibrillationTutorial

Donghyeon: usually electrodes are assumed in these models to be perfectly conducting --- so if you are using voltage boundary conditions (Dirichlet) as we do in the defib modeling you don't need an explicit electrode model except for knowing which nodes are on each electrode. If you want to set current boundary conditions instead (say, for injecting current rather than applying voltage) then the boundary conditions become a little more complicated. I don't know if SCIRun has this capability --- Jess did you guys put this in for when you model "stray leads"? I think we know how to do this in general, and it is on our list of things to do, so if it is not there and it would help you, Donghyeon, please let us know what application you have in mind and we can try to assign someone to work on it --- we will need it fairly soon ourselves I believe. If on the other hand you want your electrodes to have some non-zero resistivity, then I think Jess is correct that you have to then consider them as simply another "tissue type" --- but you would still need to define boundary conditions at the edge of the electrodes, no? hope this helps clarify things ... Dana On 10/14/11 1:35 PM, Jess Tate wrote: 970D09FE-9C54-4850-B5F0-C5C361C1EF46@sci.utah.edu" type="cite"> In our model that is in the defib tutorial, we set the entire electrode to the same potential in the boundary condition (AddKnownsToLinearSystem), so if you are using this setup directly, then the conductivity values inside the electrode are irrelevant.  That said, the space where the electrodes are also have conductivities assigned to it as dictated by the segmentation.  If for some reason you wanted to set those regions where the electrodes are to a separate than that of the tissue (segmentation), you would have to incorporate that into your segmentation somehow.  One way to do that would be to map some value representing the electrode locations onto the segmentation mesh used in the FEM calculation (1 at points for the electrode and 0 elsewhere), you could use the MapDataOntoNodes or Elements (make sure it matches the data centering for the segmentation).  You can then use CalculateFieldData to set some kind of comparison between the segmentation and the electrode label to make the electrode region the value you want, perhaps a new index and append your conductivity matrix.  The comparison would be something like    RESULT = (DATA2 == 1)*12 + (DATA2==0)*DATA1;   Where DATA1 is the segmentation indices, DATA2 marks where the electrodes are, and 12 is the new indices (the example segmentation has 11 tissues already if I remember correct).  You would then add another row to your conductivity matrix to the value you want for the electrode.   Hope that helps.   cheers, Jess   On Oct 13, 2011, at 1:05 AM, Donghyeon Kim wrote: Hi Jess i understand your question. But what i mean is, when we model electrode, we set the shape of electrode and  set the boundary condition, current density or potential on or below the electrode surface and we set the conductivity in electrode. Therefore i said that i want to set the conductivity of electrode. make sense? Regards, Donghyeon --- Original Message --- From : "Jess Tate"<jess@sci.utah.edu> To : "Darrell Swenson"<darrell.swenson@gmail.com>, danielkim@gist.ac.kr Date : 2011/10/13 목요일 오후 3:00:11 Subject : Re: [SCIRUN-USERS] set electrode conductivity on DefibrillationTutorial I don't understand why you want to set the conductivity values for the electrodes, in the example they are the sources.  the conductivity values at locations where you know the potentials (the electrodes) are not needed and redundant (if not inconsistent).     The example does provide an example of how to set conductivity values for various regions Identified by the segmentation.  this is done by using a matrix  values that correspond with the segmentation values which can be one of the inputs for BuildFEMatrix, or you can use ConvertIndicesToFieldData as in the tutorial.  Basically, you just want to map the conductivity value to region you want it to.   cheers, Jess   On Oct 12, 2011, at 10:49 PM, Darrell Swenson wrote: Sorry to forward these to you, but I am not sure you are on the list and I cannot look at them for at least a couple of days.  Feel free to answer of ignore.  If one of us does not answer, chances are that no one will. Begin forwarded message: From: Donghyeon Kim <danielkim@gist.ac.kr> Date: October 12, 2011 10:22:57 PM MDT To: scirun-users@sci.utah.edu Subject: [SCIRUN-USERS] set electrode conductivity on DefibrillationTutorial Reply-To: Donghyeon Kim <danielkim@gist.ac.kr> Hi, all I'm try to FEM simulation based on DefibrillationTutorial way, But this tutorial doesn't provide how to set electrode conductivity(can electrode), Is it impossible? Or is there a better way to simulate FEM on SCIRun? Regards, Donghyeon       --------------------------------------------------  DONGHYEON KIM, MS Student  Dept. of Info. & Comm., Gwangju Institute of Science and Technology(GIST)  261 Cheomdan-gwagiro, Buk-gu, Gwangju 500-712, South Korea Tel: +82-62-715-2266 Cell: +82-10-9361-3781 E-mail: danielkim@gist.ac.kr URL: http://biocomput.gist.ac.kr           --------------------------------------------------  DONGHYEON KIM, MS Student  Dept. of Info. & Comm., Gwangju Institute of Science and Technology(GIST)  261 Cheomdan-gwagiro, Buk-gu, Gwangju 500-712, South Korea Tel: +82-62-715-2266 Cell: +82-10-9361-3781 E-mail: danielkim@gist.ac.kr URL: http://biocomput.gist.ac.kr