Variable conductivity of white matter for epilepsy source localization

Dear all,

I am currently going through brainstorm documentation as a tool to localize the epileptogenic zone on sEEG/EcoG. Pr Karas was wondering if one can vary the white matter conductivity in the fibers close to the epileptogenic zone in brainstorm. Indeed, repeated seizures render some tracts more conductive. If it is not yet available in brainstorm and/or DUNeuro, I would like to develop it for our studies. I have experience in developing finite element codes in matlab and C++.

Thank you

Regards,

Anne-Cecile Lesage, PhD

Research scientist

University of Texas Medical Branch, Galveston, Texas

Answered by @tmedani over email:

The current implementation does not allow the change of the conductivity from the brainstorm interface/scripts.
However, this can be possible by changing portions of the code and the input files to DUNEuro.

You can do that by using the conductivity tensors defined at each FEM element.
If you can localize the area where you want to change the conductivity,
you can then find the associated FEM element and then change the local conductivity, and then call again the FEM solver.

If you want to try this, I'll have a look on the code and guide you to where you need to change it.

1 Like

dear @aclesage ,

I am working on a similar task, trying to feed the brain model solver with exact conductivities in specific area. I am wondering, have you been successful with your implementation? Would you be so kind as to share your solution?

thank you wery much

kind regards,

David

Hi @DK05

I’m also interested to hear from @aclesage.

Also, you can check the new feature here

It allows the refinement of the specific ROI of the brain and head. If you review the code, you can modify it to define this region as new tissue and then assign a different conductivity for modeling purposes.

Best,
Takfarinas

well i run into errors generating a 3d FEM source model with seeg so i postponed my exploration

I am focusing on a study using the BEM right now.

Can you post the error related to the FEM?