Back to the main page.

Bug 622 - overall peak-to-peak value calculation instead of channelwise peak-to-peak in ft_artifact_threshold

Reported 2011-04-28 15:16:00 +0200
Modified 2011-05-23 10:36:28 +0200
Product: FieldTrip
Component: core
Version: unspecified
Hardware: All
Operating System: All
Importance: P1 trivial
Assigned to: Eelke Spaak
Depends on:
See also:

Andreas Wollbrink - 2011-04-28 15:16:52 +0200

Created attachment 44 alternative function to ft_artifact_threshold.m it seems to me that in ft_artifact_threshold the amplitude rangeval is based on the minmax value accross all channels instead of calculating a peak-to-peak value per channel due to different offsets on different channels this may cause strange artefact findings attached you find a modified version of the function which serves as a replacement this function has 2 extra output parameters: - badTrials : vector list of all trials defined as bad - chnArtRate : number of BAD trials per channel divided by the total number of trials

Eelke Spaak - 2011-05-06 09:56:50 +0200

Thanks for the bug report, and for the proposed fix. I have updated the function to use your within-channel range checking.