Hi everyone and thanks for your attention.
I am analysing EEG data for a research project (I am using EEG- and ERP-Lab). Sometimes I see channels with a bad signal for only half of the trace. It seems a shame to waste so much good signal by interpolating the whole channel and I can't remove only some epochs.
I have recently started analysing EEG data, so, both theoretically and practically, I don't know if the issue can be solved by interpolating only half of the channel trace.
If you have already dealt with this issue, let me know what you think!
Thanks again,
Angelo
Hi,
Yes, you can interpolate only where the channel starts to behave bad (e.s. loose contact). Of-course an interpolation is always an estimate of what is happening underneath the EEG electrode (bad channel), and the situation is different if you have 10 channels versus 100 channels. Also, the interpolation of inner channels is more precise in comparison to the outer channels (the ones on the border).
Back to the question, if the interpolation is performed well you would expect similar time-series properties in the trace where it was still working well and where you performed the interpolation. You can also do some tests to check the difference.
Hi Angelo,
I am not sure if I have well understood your question, but if your intention is to delete epochs from some channels, you can easily do it with Brainstorm. https://neuroimage.usc.edu/brainstorm/Introduction
I hope that it could help you :)
- Badges
- Science topic
- Similar topics
- Data Mining
- Classification