Co-registration of multiple runs

1

Hi Marc,
I am interested in analysing time-frequency plots. I have multiple trials for each subject. I am planning to generate time-freq plots mostly for small clusters of sensors (size would depend upon the computational limit). As I was going through the tutorial https://neuroimage.usc.edu/brainstorm/Tutorials/ChannelFile#Multiple_runs_and_head_positions

You have three options if you consider grouping information from multiple runs:
Method 1: Process all the runs separately and average between runs at the source level: The more accurate option, but requires more work, computation time and storage.

Method 2: Ignore movements between runs: This can be acceptable if the displacements are really minimal, less accurate but much faster to process and easier to manipulate.

Method 3: Co-register properly the runs using the process Standardize > Co-register MEG runs: Can be a good option for displacements under 2cm.
Warning: This method has not be been fully evaluated on our side, use at your own risk. Also, it does not work correctly if you have different SSP projectors calculated for multiple runs.

I came across the warning for method 3. Thus, wanted to check if there has been any evaluation of it.

3

Hello,

Unfortunately, I'm not very familiar with that process. I had a quick look and if I understand correctly, it creates an interpolation of the magnetic field based on an inverse/forward model with a grid of source points inside a sphere centered in the helmet. This should be more than "good enough" for most preliminary rough exploration at the sensor level.

I'm a bit surprized that you would talk about computational limit at the sensor level. That's normally not an issue until we get at the source level with thousands of source points.

But to answer your specific question, there probably hasn't been any further testing or validation since the tutorial was written.