[Simnibs-discuss] tDCS electrode is being simulated as skin

[Oula Puonti]

Hi Gülce,

Yes exactly, you only need the center coordinate for the electrode from
the MRI scan so that the electrode is placed on the correct location in
the simulations. Simnibs will automatically place the electrode on the
skin in the simulation (by projecting it to the closest point as you
said). The only thing you need to define additionally is the gel
thickness (2mm) when you're setting the electrode shape. 

Best,
Oula
 
On Mon, 2023-01-30 at 12:11 +0100, Gülce Lale wrote:
> Hi Oula,
> 
> I just wanted to clarify: the coordinate (x,y,z) defines the location
> of the electrode in the space, therefore, the z axis does not
> determine the distance of the electrode to the skin. As it states in
> the website the coordinates do not need to be exactly in the skin
> surface, as SimNIBS automatically finds the closest point in the skin
> surface. Therefore, whatever I give as the coordinates of the
> electrode, SimNIBS will place my electrodes on to the closest points
> to the skin surface.
>        Otherwise, how can I state in the script or in the gui that I
> would like to position the electrodes on to the skin? There is a skin
> distance parameter for the TMS electrodes, however, I could not see
> that for tDCS electrodes.
> 
>        Thank you for your help in advance.
>        Best wishes,
>        Gülce
> 
> 
> > On 27 Jan 2023, at 14:55, Oula Puonti <oulap at drcmr.dk> wrote:
> > 
> > Hi Gülce,
> > 
> > You can just take the center coordinate (x,y,z) for the electrode
> > from
> > the MRI, put that in the GUI (or a script), and project the
> > coordinate
> > to the skin surface. This will keep the x and y coordinates (more
> > or
> > less) the same but set the z so that the position is on the skin.
> > After
> > that, when you define the electrode shape, just add a 2mm gel layer
> > and
> > that should be it.
> > 
> > Best,
> > Oula
> > 
> > 
> > On Wed, 2023-01-25 at 14:15 +0100, Gülce Lale wrote:
> > > Dear Oula,
> > > Thank you for the help. We were able to label electrode gel as
> > > background in the segmentation. However, when deriving our
> > > electrode
> > > center coordinates, we still define the top of the electrode gel.
> > > We
> > > know that we apply 2mm of electrode gel on top of the electrodes.
> > > Thus, the coordinates acquired from the MRI image should be 2mm
> > > below, so that they are at the actual skull height. 
> > > How do we convert length measures such as 2mm into world
> > > coordinates
> > > which SimNIBS uses? And, can I find information on this on the
> > > website?
> > > Thank you for your time.
> > > Best wishes,
> > > Gülce
> > > 
> > > > On 20 Jan 2023, at 19:11, Oula Puonti <oulap at drcmr.dk> wrote:
> > > > 
> > > > Hi Gülce,
> > > > 
> > > > You should leave them out, i.e., label them as background.
> > > > SimNIBS
> > > > places the electrodes automatically during the simulation, so
> > > > you
> > > > don't
> > > > need to add them to the headmodel yourself.
> > > > 
> > > > You can record the center coordinate of the electrodes from the
> > > > MRI
> > > > scans and pass that to the simulation so that the electrodes
> > > > are
> > > > placed
> > > > correctly.
> > > > 
> > > > Best,
> > > > Oula
> > > > 
> > > > 
> > > > On Fri, 2023-01-20 at 17:25 +0100, Gülce Lale wrote:
> > > > > Hi Oula,
> > > > > 
> > > > > Thank you very much for your response, I will try it with
> > > > > charm
> > > > > and
> > > > > SimNIBS4.
> > > > > I would like to clarify something and ask for your
> > > > > opinion. What
> > > > > tissue the electrodes should be segmented as? Should they be
> > > > > left
> > > > > out
> > > > > of the segmentation? Or be segmented as one of the other 15
> > > > > tissues?
> > > > > 
> > > > > Thank you for your help in advance.
> > > > > Best wishes,
> > > > > Gülce
> > > > > 
> > > > > > On 17 Jan 2023, at 19:54, Oula Puonti <oulap at drcmr.dk>
> > > > > > wrote:
> > > > > > 
> > > > > > Hi Gülce,
> > > > > > 
> > > > > > I would consider this a segmentation error. You might want
> > > > > > to
> > > > > > try
> > > > > > the
> > > > > > new segmentation method (charm) included in simnibs version
> > > > > > 4.
> > > > > > In
> > > > > > my
> > > > > > experience it does a bit better job on ignoring the
> > > > > > electrodes
> > > > > > in
> > > > > > the
> > > > > > type of MRI data you have.
> > > > > > 
> > > > > > The other option is that you edit the skin masks, which are
> > > > > > inside
> > > > > > the
> > > > > > MASK_prep folder, manually and then re-run the surface and
> > > > > > volume
> > > > > > masking steps.
> > > > > > 
> > > > > > The manual editing process was made easier in the new
> > > > > > version,
> > > > > > see
> > > > > > here: 
> > > > > > 
> > > > > > https://simnibs.github.io/simnibs/build/html/tutorial/advanced/fix_headmodel.html
> > > > > > 
> > > > > > so you might want to consider using SimNIBS4 in any case.
> > > > > > 
> > > > > > Best,
> > > > > > Oula
> > > > > > 
> > > > > > 
> > > > > > On Tue, 2023-01-17 at 15:50 +0100, Gülce Lale wrote:
> > > > > > > Dear SimNIBS Team,
> > > > > > > We conduct post-hoc modelling for each subject using T1
> > > > > > > and
> > > > > > > T2
> > > > > > > images
> > > > > > > acquired while the participant was in the scanner with
> > > > > > > the
> > > > > > > tDCS
> > > > > > > electrodes attached. Here SimNIBS headreco images the 2mm
> > > > > > > layer
> > > > > > > of
> > > > > > > electrode paste and includes in in the head mesh but
> > > > > > > segments
> > > > > > > it
> > > > > > > as
> > > > > > > skin (see attached images).
> > > > > > > According to SimNIBS skin conductance is 0.465 s/m and
> > > > > > > the
> > > > > > > electrode
> > > > > > > material conductivity is 29.4 s/m. One would expect the
> > > > > > > Ten20
> > > > > > > Electrode paste which is applied with a 2mm thickness
> > > > > > > between
> > > > > > > the
> > > > > > > electrode and skin to have a different measure. However,
> > > > > > > since it
> > > > > > > is
> > > > > > > being considered and modelled as skin this could results
> > > > > > > in
> > > > > > > deviations in the electric field modelling.
> > > > > > > How do we account/solve this?
> > > > > > > Could we model the electrodes 2mm below the head surface,
> > > > > > > knowing
> > > > > > > that the outer 2mm thickness of the skin are actually due
> > > > > > > to
> > > > > > > the
> > > > > > > paste? Thus, modelling the electrodes on the “actual”
> > > > > > > skin
> > > > > > > outer
> > > > > > > layer (and not on top of the paste). SimNIBS then allows
> > > > > > > us
> > > > > > > to
> > > > > > > specify past + electrode thickness. Where we assume
> > > > > > > SimNIBS
> > > > > > > calculates a different conductivity measure for both.
> > > > > > > Is there an alternative solution to best account for the
> > > > > > > change
> > > > > > > of
> > > > > > > conductance?
> > > > > > > Thank you for your help in advance.
> > > > > > > Best wishes,
> > > > > > > Gülce 
> > > > > > > 
> > > > > > > Gülce Lale
> > > > > > > MSc Neurosciences
> > > > > > > 
> > > > > > > _______________________________________________
> > > > > > > Simnibs-discuss mailing list
> > > > > > > Simnibs-discuss at drcmr.dk
> > > > > > > https://mailman.drcmr.dk/mailman/listinfo/simnibs-discuss
> > > > > 
> > > > 
> > > 
> > 
>