[FieldTrip] Reject trials based on standard deviation

Heidi Solberg Okland Heidi.SolbergOkland at mrc-cbu.cam.ac.uk
Mon Jan 22 14:15:58 CET 2018


Dear Fieldtrippers,

I'm currently working on an MEG dataset and have so far done automatic rejection of bad channels using Maxfilter and then done ICA-based artefact reduction (eyeblinks and heartbeat) in MNE Python. I now have my data in Fieldtrip trial structures. What is left is to remove trials that are outliers/noisy because the subject moved/coughed/etc. I have tried to use ft_visinspect with variance as metric, but the issue I have is that the variance is displayed in actual units (e.g. 1e10^23) rather than in terms of standard deviations. This is problematic, as my trial rejection will be based on my subjective decisions that some trials look like outliers and not others. What I would like instead is to just do something like "discard trials where the amplitude is more than X standard deviations above or below the mean", which is reproducible. Is this way of rejecting trials implemented in FT? I have seen the tutorial on automatic artefact rejection, but it looks more like something that would be useful for removing e.g. eyeblinks.

Best wishes,
Heidi


----------------------------------------------------------

Heidi Solberg Økland
PhD candidate, Language group

Tel: 01223 273721
MRC Cognition & Brain Sciences Unit
University of Cambridge
15 Chaucer Road
Cambridge
CB2 7EF

Web: http://www.mrc-cbu.cam.ac.uk<http://www.mrc-cbu.cam.ac.uk/>

Social media:
[facebook-flat-logo-01]<https://en-gb.facebook.com/MRCCBU/>   [twiiter-flat-logo-02] <https://mobile.twitter.com/mrccbu?lang=en>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20180122/5ed5f304/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.png
Type: image/png
Size: 954 bytes
Desc: image001.png
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20180122/5ed5f304/attachment-0002.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image002.png
Type: image/png
Size: 1676 bytes
Desc: image002.png
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20180122/5ed5f304/attachment-0003.png>


More information about the fieldtrip mailing list