SCIRun User Mailing List

Text archives Help


Re: [SCIRUN-USERS] Regarding Magnetic field simulation


Chronological Thread 
  • From: Petar Petrov <pip010@gmail.com>
  • To: Jess <jess@sci.utah.edu>
  • Cc: Vikas R bhat <vikasraghubhat@gmail.com>, "scirun-users@sci.utah.edu" <scirun-users@sci.utah.edu>
  • Subject: Re: [SCIRUN-USERS] Regarding Magnetic field simulation
  • Date: Mon, 18 Sep 2017 12:40:24 +0200
Hi Jess,

From my understanding the module SimulateForwardMagnetic is ill suited for TMS. Maybe I am missing something. I looked before in the example for brain-stimulation but not really convinced you are doing it right. Also, no such thing as primary current and secondary current. Thats not how electricity works, you can have multiple fields that lead to 1 current only, though complex in the case of volume conductance. Further, in the case of TMS, you dont care about J but E, unlike tDCS. Biologically TMS polarizes neurons (most likely pyramidal cells) directly, while tDCS changes the intra-cellar properties through injected currents. Thus TMS is a stimulator and modulator while tDCS is only modulator.

Further, the claim that "The computation of secondary currents by solving the FEM
problem is rather time consuming and computer simulations have shown that its impact is
rather small (compared to primary currents) on the current density J."
Have not checked Js, again cause irrelevant, but this is simply NOT true for E-fields. Anything further than 2cm from the hotspot in fact the secondary field becomes dominant !
It has been shown that you can excite the WM directly with TMS that is further/deeper than the hotspot of figure-8 coils.

I have developed my own TMS modules (1 for generation of geometry via wires and 2nd to sovle the vector potential using BiotSavart). In principle instead of wires you can use dipoles too.
There I introduce the complete vector potential from the coil on each node of the mesh, not sure why you use ROI, you should not IMO. To derive the RHS for FEM I combine the output of BuilVolRHS + BuildSurfRHS.

I need to dig a bit more, maybe compare with my results vs yours, then come back to you.

Cheers,
Petar

On Tue, Sep 12, 2017 at 12:21 AM, Jess <jess@sci.utah.edu> wrote:
Hi Vikas,

I’m not sure what your question is, but the TMS example provided will analytically calculate the magnetic field from a coil (modeled with dipoles) in the  SimulateForwardMagnetic, then convert them to current sources which are used as know in the FEM potential calculation.  You can calculate the potentials at detector positions by extracting the potentials at those locations.  You could generate a type of leadfield matrix by doing this calculation for a delta function with all sources and concatenating the results.

cheers,
Jess



On Sep 8, 2017, at 10:14 PM, Vikas R bhat <vikasraghubhat@gmail.com> wrote:

Respected Jess,

Is SimulateForwardMagnetic field module uses variational methods like FEM (discrete Maxwell equations) to generate magnetic field from the known electric field, dipole positions and detector positions?


--
Thanks,

Vikas.R.Bhat

________________________________________________
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.



________________________________________________
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.



--
All the best,
Petar Petrov
http://ppetrov.net

Archive powered by MHonArc 2.6.18.

Top of page