Back to the main page.

Bug 2899 - create template layouts for EGI nets

Status NEW
Reported 2015-05-22 13:57:00 +0200
Modified 2016-02-25 01:37:56 +0100
Product: FieldTrip
Component: plotting
Version: unspecified
Hardware: PC
Operating System: Mac OS
Importance: P5 normal
Assigned to: Anna Wieczorek-Taraday
URL:
Tags:
Depends on:
Blocks:
See also:

Robert Oostenveld - 2015-05-22 13:57:21 +0200

as discussed following last M/EEG meeting. The idea is to start from page 123 in the pdf below - which has the layouts for all the caps: http://ganesha.uoregon.edu/images/8/8c/Gsn_013107.pdf


Robert Oostenveld - 2015-05-22 14:19:42 +0200

On 21 May 2015, at 18:31, Anna Wieczorek-Taraday wrote: Please find attached: EGI_GSN_64_v1.mat - the layout created today, according to our short chit-chat. I did not include mask. I also did not change the labels so far. There are also two electrodes ('comnt' and 'scale'), which pop out automatically and were not present in the original layout. EGI_GSN_64_v1_layoutplot - the plot obtained from fieldtrip. Beatmaps for all the EGI caps. The layout file that I used before for topoplotting ('old_layout.mat'). ----- I copied your files to /home/common/matlab/fieldtrip/data/test/bug2899 where you can find them under version1. I also made a version2 for the screenshots in higher resolution (300dpi of the full page), which are also there. For the first one you already did it is ok to stick with your screenshot, for the others I suggest you use mine (because they are of better quality). If it is fast, you could redo the first one as well. I recall that the EGI system all have channel names that start with E, i.e. E1, E2, etc. That is how they are coded in ft_senslabel and how ft_senstype detects them. So please


Robert Oostenveld - 2015-05-22 14:36:45 +0200

Once all locations have been digitised from the bitmaps and given the correct name, the next step is to make outlines and masks. I will attach here on the bugzilla page a first attempt to sketch an outline in Adobe Illustrator (which I hope you also have on your PC). In this one it is easy to scale the head shape and move it around until it is "visually pleasing". One a useful size and shape is clear, you can then take the head shape we already have in the default layouts and scale+shift it to get it approximately the same as in AI. Note that my attempt is only version 1. You know the nets better than I do, so you should take the head shapes in the AI file and update them to your liking. It could well be that I made the heads too large compared to the electrode nets. The mask is used for interpolating, and could be a single circle that extends outside the outline (i.e. goes below the nose and ears), as sometimes used in EEGLAB like here http://sccn.ucsd.edu/eeglab/comp252.jpg Please let me know if you have questions.


Robert Oostenveld - 2015-05-22 14:38:19 +0200

the files are too large to attach here. Please get them from https://dl.dropboxusercontent.com/u/3206396/egi_outline_v1.ai https://dl.dropboxusercontent.com/u/3206396/egi_outline_v1.pdf


Robert Oostenveld - 2015-05-22 14:39:21 +0200

let me add Joseph Dien to the CC, he uses EGI as well and might want to share his opinion on the EGI template layouts.


Robert Oostenveld - 2015-05-22 14:40:36 +0200

we will also need a consistent naming scheme that corresponds to the EGI documentation. We should revise the naming scheme for the SFP files as well, so that it is clear which SFP file corresponds to which layout (and page in the PDF documentation).


Anna Wieczorek-Taraday - 2015-05-25 14:26:11 +0200

(In reply to Robert Oostenveld from comment #2) Hi Robert, I created digitised layouts, there are in the folder: 'H:\common\matlab\fieldtrip\data\test\bug2899\layfiles'. Now I do not quite understand what you want me to do with masks and oulines. Would you like me to import them from the .ai files? Or to copy such masks from default layout files that you use in fieldtrip? If yes, where can I find these? best Anna


Robert Oostenveld - 2015-05-26 09:03:08 +0200

(In reply to Anna Wieczorek-Taraday from comment #6) My idea with the AI file was to use it to play around with the outline+mask by just shifting/scaling it, as it is easy there with the user interface and mouse. Once the outline+mask look nice (and consistent) in AI for each of the caps, my idea was to use that figure (in AI on the computer screen) as example for making the outline+mask in MATLAB. So the idea is to use AI to reduce the amount of time you need in MATLAB to make it look visually appealing.


Jan-Mathijs Schoffelen - 2016-02-22 09:40:22 +0100

what is the status of this?


Joseph Dien - 2016-02-25 01:37:56 +0100

I tried to look at Robert's ai and pdf files but they are no longer on dropbox. Anna's files are not accessible to me. Joe