time for this was changed or the GMT+1 is wrong ? Thank you in advance
it was written 17:00 GMT+1
Hi this is Cristina, and I haven´t used yet
Hi, I'm Giulia. My research group is using it, but I have not used it directly
Hi, I'm Vikram Shenoy (Assoc. Research Scientist) at Kessler Foundation (New Jersey). I have used ROAST just once before for testing the modeling using individual MRI
I am more interested in learning how to use ROAST for reporting in scientific writing. I have an ongoing project on HD-tDCS for stroke and TBI rehab.
Chun Hang Eden TI
hi I am Eden from Chinese University of Hong Kong. I have used ROAST once to generate head meshes. I am more interested to know more about using ROAST on MRI images with pathologies like stroke
is the problem because of lesion mapping?
Hi, I'm Adri from the University of Szeged and I have not used ROAST before; however, I have tried other simulators (e.g. SimNIBS). I do research applying tDCS and tACS so I would be interested to model the effects of the montages I use.
Hello thank you for the session. I haven't used it either. I would be interested to learn about applications in cognitive neuroscience, esp. top-down processes, cognitive control etc
Hi I’m Weronika, I’ve used SimNIBS and haven’t used ROAST yet. I’m wondering if it is possible to simulate high frequency tES such as tRNS?
yes, i will be sure to discuss this later
or maybe now
is it possible to model montages with extracephalic electrodes?
Can we use individualized electrode positions? digitized using 3D digitizer?
Chun Hang Eden TI
Can we used customized electrode location?
Can you use a template MRI if you don’t have the subject’s MRI?
is that MNI coordinate system?
What coordinate system is to be used?
is it possible to perform intra-cerebral stimulation simulations?
What about the size of the electrodes? Is it possible to define the area (cm2 or mm2) of electrodes?
we can specify current for the simulation but I dont think we can specify frequency of the current. so can I say that tDCS and tACS simulations for same electrode placement will give same output?
Following up with the above question, I read from a recent paper that for tACS delivered over two locations, the electric field distribution would be different depending on the relative phase of the 2 signals. How do we input that information?
what we should consider if we use the MRI imaging from the patient such as epilepsy with depth electrode?
does it matter if you have more than one current source? For example, if I use 2 different stimulators instead of one multichannel stimulator, to input current to >2 tDCS electrodes?
How to transform electric field maps from native space to MNI space for group level analysis?
are pads always presumed to be rectangular?
It takes in account if we are using gel or saline based electrodes?
Hello. Can I have an arbitrary shape of electrode? e.g. an ellipse?
It was suggested that we wait with questions until Andy had time to show more material. I will bring them up at the end.
Which one is better? Surface-based models or volume-based?
sorry i dont have any voice
non-uniformty under the electrode: yes
If there is a segmenting error when the program runs, is there a way to go back and fix it?
Is it possible to import a mesh created outside of ROAST? can manual corrections be done to the segmentations?
Thanks for the great talk Andy! I was wondering if you could talk about how to interpret the vector information produced by ROAST? For example to compare current direction between subjects? How could I tell which direction current is flowing in with respect to the cortical surface, just from looking at the numerical outputs in ‘ef_all’?
I often see warnings (not errors) about a list of nodes not found/exist when running ROAST, how worry should I be about that?
Regarding the conductivity, do you make any adjustment upon changes in frequency or temperature?
Sorry I didn’t quite catch what you said, is gel included by default, or do you add this in?
One example of an error we received was that the segmentation incorrectly perceived the eyes, which created holes in the skull.
(Came in tad late, sorry.) Could you provide an overview of numerical infrastructure -- FDM/FEM, default solver, etc.? Does user have control over these?
you mentioned empirically reliable conductivity values. do you refer to this study: Y. Huang, A. Liu, et al., Measurements and models of electric fields in the in vivo human brain during transcranial electric stimulation, eLife 2017;6:e18834 ?
yes, that is a good paper to read up on conductivityand validation
yes, gel is included by default
What kind of MRI scans would you recommend to use for the simulations? Especially in regard to possible segmentation errors (e.g. in some areas wrong bone segmentation). Would you recommend to use e.g. only one T1 MRI scan, T1 and T2 MRI scans or T1 with fat suppression?
A nice review paper on Electrical conductivity values in the brain is : Variation in Reported Human Head Tissue Electrical Conductivity Values by McCann et al. (The MREIT values are not MREIT), but for the rest it's a good overview paper
standard T1 used for brain imaging is good
sorry if i missed this, if we use a rubber electrode covered by a sponge, is this default or do we need to add the sponge?
T2 helps, but not really needed
What value do you recommend to do this prophylactic "zeropadding"
not sure about fat saturation
zeropadding, enough to fit the electrode.
It would generate optimal montage out of which montages (regarding eilec size, shape, position, current, etc.)?
Is there any way to simulate the electric field intensity as a function of time? e.g. At the beginning of the stimulation vs. 20 min. post stimulation?
Can you say a bit more about what the leadfield is and how roast_target uses it to find an optimal montage?
I did'n understand if it gives you the best solution or you have to specify the number of electrodes.
there is no reason to expect that fields would change in time
Does defacing of structural MRI affect electric field modeling?
there are no secific studies that i am aware of how much it would change, but in theory it will have an effect.
its best to use the original MRI is possible.
the "lead field" is the same lead field of EEG source localization, it relates currents on the electrode with fields generated in the brain, and vise versa, currents in the brian and voltages on the electrodes
"it gives you the best solution or you have to specify the number of electrodes" ...when you optimize for intensity in a given target, you will always get the minumum number of electrodes possible.
when optimizing for focality, then you will get often as many electrodes as you are allowing it to use.
which is limited by your hardware.
Does ROAST give you control over shape and mapping order of elements?
Thanks so much, this was really helpful
Will we have access to the recording of this presentation?
yes, we will post on the ROAST website
Thank you for showing all features of ROAST. What are future additions to the program you would like to do?
infrastructure for lesion delineation?
thanks for great presentation
I know I mentioned this earlier, but is there a way to resolve segmentation issues (ie. Holes in the brain) if this problem arises
that's a really GOOD long term plan!
Is pediatric modeling possible?
Thanks for the great presentation
which is the infrastructure about lesion delineation?
do the parameters for the MRI acquisition matter? or can it handle anything. thanks for a great presentation!
thanks for the session and all the useful information..
look for "MultiPrior" on pulbications are Parralab.org
that takes care of segmenting lesions
we plan to incorporate into ROAST
I will be talking about the lesion segmentation on Wednesday's first session
did you say Deepmedic or deepscan for lesion segmentation?
thanks @Lukas Hirsch.. I look forward to attending your talk.
MultiPrior, or come to the presentaiton of Lukas Hirsch
Thank you so much for this presentation!
thanks for the great presetnation :)
Thank you! Great talk and great software
Thank you very much!
Thank you very much!
Thanks ! Great talk !
Thank you for the informative presentation!
Great talk. thanks!