Artefact removal without cardiac and VEOG channels

Discussions
,
1

Hello BST,

  • I have large amounts of clinically acquired Resting State MEG data. Unfortunately, it doesn't contain ECG and VEOG channels to use SSP as described in tutorials.
  • The other way is to use ICA, which I tried, but it is becoming increasingly difficult as I have to visually examine all the ICA components for every patient and remove them.
  • Alternatively, The process " Artifacts > Detect other artifacts " identifies epochs of time that contain typical artifacts from eye movement, subject movement or muscle contractions. Can I use this procedure (Artefacts > Detect Artefacts). Is this procedure as good as SSP with ECG and VEOG channels?
    Any advice is very much appreciated as I am currently struggling how to remove these artefacts.
    Regards,
    Paddu.
2

The process "Detect other artifacts" is not well adapted to detect blinks and cardiac peaks.

For the ocular artifacts: you should be able to detect the blinks on a frontal MEG channel instead of a VEOG channel. It might require some tuning of the detection parameters (if the process "Detect eye blinks" does not work well, try using "Detect custom events" instead):
https://neuroimage.usc.edu/brainstorm/Tutorials/ArtifactsDetect#Detection:_Custom_events

The heartbeats are more complicated to detect, because the peaks of amplitude are most of the time not sticking out clearly from the rest signal recorded by the MEG sensors.
On some subjects that have a lot of cardiac artifact, you might still be able to detect them on a temporal MEG channel using the "Detect heartbeats" or "Detect custom events" processes.
Alternatively, either you don't correct for the heartbeats, or you mark a few tens of cardiac peaks manually.