help > Template MNI brain and subcortical overlay does not match with electrode positions
Showing 1-5 of 5 posts
Apr 21, 2021 11:04 AM | lucakolibius
Template MNI brain and subcortical overlay does not match with electrode positions
Hi,
I really like the visualisation and the intuitive design of Surfice!
(1) So far I have been using the MNI brain template provided in \surfice_windows\sample\mni152_2009.mz3
(2) For the hippocampal overlay I have used \subcortical_mz3\Right-Hippocampus.mz3 from https://brainder.org/research/brain-for-blender/ as suggested in the wiki
(3) The electrode positions are from various patients implanted with sEEG and are all in MNI space. Based on visual inspection in native space and when plugging in the MNI coordinates into https://bioimagesuiteweb.github.io/webapp/mni2tal.html the electrodes are in the hippocampus
However, when overlaying everything none of the electrodes are in the hippocampus (see misplacedElectrodes.png, left). I have used another subcortical overlay (thalamus and CA1) and the mismatch between the overlay and the template brain is even bigger (see attached, right).
The brain template from brainder.org also seems to be much smaller than the one provided in the surfice sample folder. Is this mismatch issue due to a different resolution? If so, can this be fixed (inside Surfice)?
Thank you in advance.
All the best,
Luca Kolibius
I really like the visualisation and the intuitive design of Surfice!
(1) So far I have been using the MNI brain template provided in \surfice_windows\sample\mni152_2009.mz3
(2) For the hippocampal overlay I have used \subcortical_mz3\Right-Hippocampus.mz3 from https://brainder.org/research/brain-for-blender/ as suggested in the wiki
(3) The electrode positions are from various patients implanted with sEEG and are all in MNI space. Based on visual inspection in native space and when plugging in the MNI coordinates into https://bioimagesuiteweb.github.io/webapp/mni2tal.html the electrodes are in the hippocampus
However, when overlaying everything none of the electrodes are in the hippocampus (see misplacedElectrodes.png, left). I have used another subcortical overlay (thalamus and CA1) and the mismatch between the overlay and the template brain is even bigger (see attached, right).
The brain template from brainder.org also seems to be much smaller than the one provided in the surfice sample folder. Is this mismatch issue due to a different resolution? If so, can this be fixed (inside Surfice)?
Thank you in advance.
All the best,
Luca Kolibius
Apr 21, 2021 11:04 AM | Chris Rorden
RE: Template MNI brain and subcortical overlay does not match with electrode positions
You need to be very careful that the coordinate system you use
matches the atlases you use. For example, SPM normalizes brains to
have an average size, whereas the MNI template is larger than
average. This is shown in Figure 1 of Horn et al. (2017)
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6050588/
I would suggest you consider using Lead-DBS which integrates Surfice
https://www.lead-dbs.org
MRIcroGL and Surfice will work with any coordinate system, but you need to be careful to choose the same coordinate system for all stages of your processing and visualization.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6050588/
I would suggest you consider using Lead-DBS which integrates Surfice
https://www.lead-dbs.org
MRIcroGL and Surfice will work with any coordinate system, but you need to be careful to choose the same coordinate system for all stages of your processing and visualization.
Apr 21, 2021 12:04 PM | lucakolibius
RE: Template MNI brain and subcortical overlay does not match with electrode positions
Thank you for your help, Chris! Figure 1 from Horn et al. (2017) is
very informative as well.
Is there a way to transform SPM MNI coordinates (so smaller brain size; those are used to localise our Behnke-Fried Electrodes) to an arbitrary Atlas such as the subcortical system or the provided MNI brain? In a best case scenario I would also use that approach to verify that the subcortical system and the MNI brain have the same coordinate system. Or is my best bet to localize all electrodes again in the lead-dbs framework?
Is there a way to transform SPM MNI coordinates (so smaller brain size; those are used to localise our Behnke-Fried Electrodes) to an arbitrary Atlas such as the subcortical system or the provided MNI brain? In a best case scenario I would also use that approach to verify that the subcortical system and the MNI brain have the same coordinate system. Or is my best bet to localize all electrodes again in the lead-dbs framework?
Apr 21, 2021 01:04 PM | Chris Rorden
RE: Template MNI brain and subcortical overlay does not match with electrode positions
Outside my expertise. Unless you want to spend a lot of time
validating this, it seems like lead-DBS provides a turn key
solution to your problems.
Apr 21, 2021 03:04 PM | lucakolibius
RE: Template MNI brain and subcortical overlay does not match with electrode positions
Got it! Thank you for your help and time!