Back to the main page.

Bug 1479 - reading in polhemus locations; yokogawa warning and extra .chanpos field

Reported 2012-05-16 15:04:00 +0200
Modified 2019-08-10 12:02:52 +0200
Product: FieldTrip
Component: fileio
Version: unspecified
Hardware: PC
Operating System: Windows
Importance: P3 normal
Assigned to:
Depends on:
See also:

Johanna - 2012-05-16 15:04:08 +0200

I'm loading Polhemus sensor positions by: elec=ft_read_sens(['f01.pos']); It works, and the output 'elec' is: chanpos: [67x3 double] elecpos: [67x3 double] label: {67x1 cell} unit: 'cm' But 2 issues I see: 1) .elecpos and .chanpos are redundant fields, and I'd say .chanpos should not exist in this case. It is set in ~/fieldtrip/fileio/private/ft_datatype_sens.m on lines 108-109. But it causes no errors further on, so perhaps not a problem? But as .chanpos and .elecpos are sometimes used to distinguish between EEG or MEG datatypes, then perhaps an issue there? Why have the redundant fields? 2) Just prior to those lines, on line 107, channelposition is called (~/fieldtrip/fileio/private/channelposition.m) within that function on line 43, if .ori does not exist, it is forced to exist as NaNs. That may be fine, except then on line 46 of channelposition.m, ft_senstype is called, which then thinks this 'sens' structure 'isgrad' since the sens.ori field exists. This further leads to the warning that 'could be Yokogawa system'. Suggestion to do either: A) Not add the .ori NaNs in channelposition.m (though perhaps it is needed further on?) Or, B) make ft_senstype more robust to check for non-NaN .ori rather than just presense of .ori, before setting isgrad=1.

Johanna - 2012-05-16 15:52:59 +0200

JM commented to me that if I use ft_read_headshape instead, this is not a problem, which is true. Then I can set shape.pnt to be elec.elecpos and all is fine. However, I wonder if my questions are still valid in general, ignoring that it is a .pos Polhemus file that I'm reading in?

Jan-Mathijs Schoffelen - 2013-12-12 12:57:33 +0100

I don't know whether the points Johanna raised are still an issue. Since reporting this bug I think that some things may have changed in addressing isgrad/iselec, so this may be past its best-before-date. Also, in yesterdays (2013-12-11) FT-meeting it was discussed to adjust the way fieldtrip deals with elec-structures, and to allow for data structures to have both an elec and grad field. Long story short, probably after this is implemented the current bug will become invalid anyway. Since I don't expect anyone to spend time on this issue before that, I suggest to set this one to wontfix.

Johanna - 2013-12-19 17:00:02 +0100

sure, sounds fine to me.

Robert Oostenveld - 2019-08-10 12:02:52 +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