Back to the main page.

Bug 3167 - New problems in ft_plot_sens

Reported 2016-07-23 00:15:00 +0200
Modified 2016-11-29 09:00:29 +0100
Product: FieldTrip
Component: plotting
Version: unspecified
Hardware: PC
Operating System: Windows
Importance: P5 normal
Assigned to: Jan-Mathijs Schoffelen
Depends on:
See also:

Vladimir Litvak - 2016-07-23 00:15:43 +0200

After the latest FT update plotting Neuromag sensors in SPM stopped working. Try figure;hs = ft_plot_sens(sens, 'chantype', 'megmag') with Neuromag grad. First hs is not generated and the function crashes. After fixing that, it still doesn't work because direction is all nans. Vladimir

Jan-Mathijs Schoffelen - 2016-08-08 13:17:37 +0200

confirmed. I can solve the second problem (with the nans), but I am not sure how to address the missing output argument. In the absence of coils being drawn, hs reflects a (single) handle to the set of points (as obtained by a call to plot3). When the coils are drawn, the subfuncion plotcoil is used, which creates a patch-object for each of the coils, hence an array of handles. How does SPM rely on the output argument of this function? can it handle an array of handles, or does it require a single handle?

Vladimir Litvak - 2016-08-08 13:38:41 +0200

We change the colour. Yes return an array and I'll make sure we can do something with it. Vladimir

Jan-Mathijs Schoffelen - 2016-08-08 13:48:47 +0200

OK, implemented with pull request 203. Note that when plotting the sensors as points hs is a handle to a lineseries object, and when the sensors are plotted as coils hs is an array of axis handles

Jan-Mathijs Schoffelen - 2016-08-09 09:45:46 +0200

According to me, this can now be closed. @Vlad: agreed?

Vladimir Litvak - 2016-08-09 09:48:04 +0200

I can't check now. Will reopen if necessary.

Jan-Mathijs Schoffelen - 2016-08-09 11:06:19 +0200

Oh, and isn't there an option in ft_plot_sens to change the color? Should we look into this?

Vladimir Litvak - 2016-08-09 11:08:18 +0200

Yes, that would be helpful. We like to make the sensors green

Jan-Mathijs Schoffelen - 2016-08-11 13:13:32 +0200

(In reply to Vladimir Litvak from comment #7) I think the option 'edgecolor' is intended to determine the color.

Vladimir Litvak - 2016-09-06 17:15:15 +0200

(In reply to Jan-Mathijs Schoffelen from comment #8) Hi Jan Mathijs, I looked at the current code and I see that although it makes quite nice plots for Neuromag and CTF, for EEG and just magnetometers the only thing you can get is a small black dot and that looks quite ugly. Isn't there a way to at least have something like a sphere as with the previous code? Vladimir