Hello everyone,
I am currently attempting to compute the forward head model for two subjects. For both, I am getting the following error message (below). I then checked the channel file and checked my MRI registration (also below), and I did not see where the sensors were that were "inside the brain volume". Furthermore, I was not able to refine using head points for either subjects, as I did not have enough digitized head points to perform automatic registration. Should I perform manual registration at this point?
Thank you for your help!
Jennifer
I was not able to refine using head points for either subjects, as I did not have enough digitized head points to perform automatic registration. Should I perform manual registration at this point?
If you don't have head points, and the basic registration based on the fiducials (NAS/RPA/LPA) gives reasonable results (as you screen capture shows): I don't recommend you try to do anything else.
I am currently attempting to compute the forward head model for two subjects. For both, I am getting the following error message (below).
Is this MEG+EEG recordings?
If it is, then your EEG electrode positions are not set, you should fix that.
If you don't have any EEG in the recordings: edit the channel file and set to something else than EEG the channels that might be classified as EEG.
If my comments do not help you, please share your channel file (the channel.mat): upload it somewhere and post the download link here.
Thanks
This has occurred on multiple MRIs, but no channels are listed as EEG. Here are the channel files for the 2 MRIs that I am getting this error.
Thank you,
Jennifer Meyer
channel_ctf_acc1.mat (225.9 KB)
channel_ctf_acc1.mat (194.2 KB)
If this is only MEG, then it is really your registration based on the fiducials that is incorrect.
We can't see on your screen capture, but I guess from the other side of the head, you can observe the yellow helmet surface cross the head surface? Example:
If the registration between the MEG and the MRI is obviously wrong, when you don't have extra digitized head points, there are two possibilities:
- You did not place the fiducials in the MRI in a way that matches the way they were defined in the MEG recordings (if you used the Brainstorm digitizer, you need to click the anatomical NAS/LPA/RPA points, if you didn't you need to position the head positioning coils)
=> This can be fixed: re-import the MRI and set correctly the NAS/LPA/RPA points in the MRI.
- The head positioning coils where not localized correctly by the MEG acquisition system (something went wrong during the digitization of the points, or the MEG acquisition system did not localize them correctly).
=> This cannot be fixed: the position of the head in the MEG was not properly recorded, there is no way to fix it or compensate for this missing information. You can move the sensor array to a different position in Brainstorm with the menu MRI registration > Edit, at your best guess, but this is arbitrary and you will have no way to validate that this manual fix is correct.
