Problem with computing sources and extracting time-scout-series

Bugs
1

Hi,
I'm trying to extract time-scout-series but it is not working for 5 recordings out of 60. In these 5 recordings, I always get an error and I think all numbers are NaN numbers. The signal I'm using looks good and I do the same thing for all recordings but it's not working for these 5. Do you have any idea what could be causing this?
Ps. I have the newest Brainstorm update

Best regards,
Linda

Screenshot 2022-01-26 at 19.57.42
Screenshot 2022-01-26 at 19.57.421756×540 71.4 KB

2

This is what the signal looks like but the problem is probably not the time-scout-series but when I compute the sources as it shows "Invalid scale".

image
image1920×1200 87.6 KB

3

My guess is that there is something wrong with the forward model (head model) or the source model (minimum norm): one of the two files contain NaN values. This is not causing any obvious problem until you try to plot scouts time series.

Try to understand if the problem stands in the forward or the inverse model, by inspecting the values in the models.
If the problem is the forward model: double-check the surfaces, the sensor positions, etc. You can try to compute a different head model.
If the problem is the inverse model ONLY (nothing wrong in the forward model): this might be due to problems in the noise covariance file. You need to inspect this file as well.

If you can't figure out what the error is, we would need an example dataset in order to debug it:

  • Right-click on the subject folder > File > Duplicate
  • In the duplicated subject anatomy: If you are using an individual anatomy processed with FreeSurfer: delete all the files except for the MRI, the head surface, the BEM surfaces if applicable, and the cortex surface used for the source estimation
  • In the functional side of the protocol, for the duplicated subject: Keep only one folder, including only: the channel file, the noise covariance, one imported epoch.
  • Re-compute the forward model
  • Re-compute the inverse model
  • Make sure you can reproduce the error with the source display (Invalid scale) and scout time series (error)
  • Right-click on the duplicated subject folder > File > Export subject
  • Upload the .zip file somewhere and post the download link here