Back to the main page.

Bug 1701 - ft_senstype fails to detect CTF275 when it is called on a subset of channels

Reported 2012-09-06 13:41:00 +0200
Modified 2019-08-10 11:56:29 +0200
Product: FieldTrip
Component: core
Version: unspecified
Hardware: PC
Operating System: Linux
Importance: P3 normal
Assigned to: Diego Lozano Soldevilla
Depends on:
See also:

Roemer van der Meij - 2012-09-06 13:41:32 +0200

In ft_databrowser, just before plotting, several scaling cfg-options cfg.megscale/cfg.eegscale/etc are applied. Which channels to use for e.g. the MEG option is determined ...ft_channelselection('MEG',channelspresent)... Ft_channelselection calls ft_senstype, such to determine which channels are MEG channels. When a subset of channels is selected, i.e. see bug 778, ft_senstype fails to detect the CTF275 system @line237: mean(ismember(ft_senslabel('ctf275'), sens.label)) > 0.8 In this example 57 channels are selected, leading to many zeros for the ismember, leading to mean to be below 0.8 ---> not-ctf275 How should we handle this? Use the grad as input to ft_channelselection in the databrowser? Lower the ft_senstype threshold? Determine the mean-ismember in reverse, i.e. check whether each inputchan is part of the set, instead of each setmember being part of the inputchan? CC: JM, Jorn (you worked on senstype a lot no?), and Diego (suffering from the bug atm)

Roemer van der Meij - 2012-09-06 13:44:26 +0200

*** Bug 778 has been marked as a duplicate of this bug. ***

Jan-Mathijs Schoffelen - 2012-09-12 13:47:43 +0200

Discussed in FT-meeting 12/09: add optional third input to ft_channelselection (as key-value): 'senstype', xxx so that the ft_senstype is not always necessary.

Diego Lozano Soldevilla - 2012-11-06 10:42:02 +0100

(In reply to comment #2) Hi Roemer, Don't you mind if I take you this bug? I'm in a bit rush to solve this problem to continue with my analysis Diego

Roemer van der Meij - 2012-11-06 10:56:41 +0100

Hey Diego, Sorry! I've been swamped with work the past weeks and haven't gotten to it yet, my buglist is missing me :). I was planning to do a catchup session this week, but if you're in a rush today, feel free to take it.

Diego Lozano Soldevilla - 2012-11-06 11:18:51 +0100

(In reply to comment #4) No worries! we can divide the work ;)

Diego Lozano Soldevilla - 2012-12-10 15:53:28 +0100

test script is added: test_ft_channelselection

Robert Oostenveld - 2019-08-10 11:56:29 +0200

This closes a whole series of bugs that have been resolved (either FIXED/WONTFIX/INVALID) for quite some time. If you disagree, please file a new issue describing the issue on