Hi, I got an error when I tried to run OpenMEEG for computing a head model.
Could you help me to fix this bug?
I'm using 64-bit Windows7 and MATLAB R2015b.
Microsoft Visual C++ 2010 Redistributable Package is already installed on my pc.
Thanks,
Sangtae
hi,
the OpenMEEG code does not like when you have a space in your user name (file paths) unless you protect the file paths with quotes.
See this fix:
once merged just update brainstorm and it should work.
Hope this helps
Alex
The update from Alexandre has been merged.
Please update Brainstorm and try again.
Thanks Alex and Francois, it is working now !
Hi all,
I am having the same problem with my head modeling, unfortunately. I use Windows 8, 64 bit. I installed the latest version of the Brainstorm yesterday and downloaded OpenMEEG manually as it was recommended here (https://neuroimage.usc.edu/brainstorm/Tutorials/TutBem#Errors) and installed it from Brainstorm again but it didn't work too. Microsoft Visual C++ 2010 Redistributable Package was installed as well. There is pretty much memory available in the pc at the moment. There is no space within the folder path that might cause any problem as it was indicated here (OpenMEEG bug for head model). So I would really appreciate if you could share your opinions on what might the reason that I am getting this error message, please. I attached a screenshot here but it is pretty the same with the initial message in this thread.
Thank you very much for any advice in advance,
Isil
Sorry I donât see the error message? Does it say too unknown file format?
Hi Alexandre,
thank you for your reply. Sorry for the image, so I uploaded it to Dropbox, hopefully you could find it here https://www.dropbox.com/s/98qo0uw23j2hl1e/openMEEG_error.png?dl=0
Actually it says âPlease consider updating your geometry file to the new format 1.1 âŚâ I downloaded the latest Brainstorm version, I believe it shouldnât be a problem isnât it.
Do you think would this indicate to a lack of memory?Because all process work without a problem if I decrease the number of the vertices in BEM surfaces to the lowest number (642) or run OpenMEEG on each layer separately I donât get any problems on running whole modelling. So it sounds like I am having trouble with the memory, doesnât it? Do you think making memory adjustments in Matlab would help?
Thank you very much in advance,
Isil
yes it should a memory issue. I cannot do much about it.
Alex
@Isil_P_Bilgin Make sure you didnât select a high-resolution cortex surface: the one selected in green should be a 10000V or 15000V surface
Otherwise, it is OK run to OpenMEEG with a lower number of vertices, eg. 1082/642/642, as suggested in the EEG&Epilepsy tutorial: https://neuroimage.usc.edu/brainstorm/Tutorials/Epilepsy#Head_model
Hi @joerlema
Your error is different.
It seems the surface 3 that you are using is not closed.
this is displayed on the error panel, can you correct this surface, and run it again?
Hello @tmedani
Thank you for your response. I am afraid I have no idea what is meant by "surface 3", where I can find the error panel or how I can correct this surface.
Sorry if this is a basic question!
Hi @joerlema
You are using these surfaces to complete the BEM head model.
In the error that you shared, there is a mention of surface 3.
This is probably the surface of the scalp.
How did you compute these surfaces of your model?
Can you plot and share some view, rotate it and check if something is wrong?
You can also remove and recompute them with one of the available option within bst.
Hi @tmedani
I computed a BEM model based on the T1 of the individual subject (so I didn't use a template). The following surfaces are selected:
the scalp surface looks bad, the BEM computation assumes a smoothed surface, and your scalp surface is not, this is due to the noise on the MRI.
I'll recommend smoothing it (right-click on the surface, ==> less vertices, then check how the surface looks like).
If you still can see the weird shapes on the back you may need to reconstruct the surfaces using an other method.
