SCIRun User Mailing List

Text archives Help


Re: [SCIRUN-USERS] Units used for build FE Matrix


Chronological Thread 
  • From: jess <jess@sci.utah.edu>
  • To: Douglas Wallace Todd <toddd@email.arizona.edu>
  • Cc: scirun-users@sci.utah.edu
  • Subject: Re: [SCIRUN-USERS] Units used for build FE Matrix
  • Date: Tue, 30 Mar 2021 22:06:34 -0600
  • Ironport-sdr: Bta7RAcMQEGP9onJP+nOR9dyzapND7oqaWogM37YyQSLfdSL7xGg9Q1/RPhmFIS/bBCIXA1WNl czD9EOLuokFinyZsjs1j42gmBR//lafOwYdpaH4PplCfbjtUvrdxBVnGq+cjQKIbwjBySwfLZn ZedeJKZDnxiPAO0qYDqGNB+Tf39EFtA84KuPS0CdhZgddL3Dk8C3n0zpmc1wcMPHfV9BN0wdyK HSx7No0OW8BUSiLlVjdI+oZFY01jUk6vVQEBJqnBe3X7z6nR+MQT4W2UQASVwXUydKYBDu2pQq Rt8=

Hi Douglas,

When using the AddKnownsToLinearSystems module like you are, it is adding
known voltages to the medium you are simulating. This is a way to implement
Dirchlet boundary conditions into your FE model.

To apply a current source, you will need a vector for the current densities
at each node of your FE mesh and set it as the right hand side input to the
SolveLinearSystem (it can also be used with know voltages with
AddKnownsToLinearSystem). ApplyFEMCurrentSource is a module that may help
make the RHS vector.

The FE calculations are agnostic to units, so whichever units you use are the
ones that will come out. However, practically speaking, the conductivities
should be in a unit similar to seimens/meter, or something similar, with the
length unit matching that of the mesh. If your mesh positions are in mm, then
you should use seimens/mm.

For more examples on current and voltage simulation, checkout the
FowwardInverseToolkit (https://github.com/SCIInstitute/FwdInvToolkit) and the
BrainStimulator toolkit (https://github.com/SCIInstitute/BrainStimulator). I
also hope that you have checked out the defibrillation tutorial
(http://www.sci.utah.edu/devbuilds/scirun_docs/DefibrillationTutorial.pdf).

cheers,
Jess





> On Mar 30, 2021, at 5:16 PM, Douglas Wallace Todd <toddd@email.arizona.edu>
> wrote:
>
> I have a couple of questions on the units used by the "BuildFEMatrix" and
> add unknowns to the linear system. In my system, I import a label map as an
> NNRD file and clip the exterior regions of the NRRD file. The NRRD file
> identifies skin, scalp, and a couple of brain lobe regions from a synthetic
> model of the head.
>
> I am using the "create matrix" and the BuildFEMatrix functions to --
> hopefully-- replace the labels in the label map with the conductivities
> identified by the matrix in the 'create matrix' entries.
>
> I am using a combination of CreateMatrix and Convert Indicies to filed data
> to try to generate inputs for "add Knows to Liner system". Should these
> values be expressed as voltages or current inputs, and what are the units?
>
> Is there a way to use specify a current for the stimulus inputs?
>
> I have attached a picture of the part of the network that I am concerned
> with.
>
> What are the proper units for the conductivity matrix?
> What are the proper units for the conductivity matrix?
>
> Thank you very much
>
> Douglas Todd
>
> <Screenshot from 2021-03-30 16-14-38.png>
> ________________________________________________
> SCIRun users mailing list: scirun-users@sci.utah.edu
> http://www.sci.utah.edu/software/scirun.html
> To unsubscribe, email sympa@lists.sci.utah.edu with the "unsubscribe
> scirun-users" in the message body.




Archive powered by MHonArc 2.6.18.

Top of page