Back to the main page.

Bug 2602 - document how to deal with differences in mouse head size

Status CLOSED WONTFIX
Reported 2014-06-05 21:43:00 +0200
Modified 2019-08-10 12:32:19 +0200
Product: FieldTrip
Component: documentation
Version: unspecified
Hardware: PC
Operating System: Mac OS
Importance: P5 normal
Assigned to: Chung Ki Lee
URL:
Tags:
Depends on: 2603
Blocks: 2635
See also: http://bugzilla.fcdonders.nl/show_bug.cgi?id=2604

Robert Oostenveld - 2014-06-05 21:43:04 +0200

I have made a start with documenting this on http://fieldtrip.fcdonders.nl/development/mouse_eeg. It would be better to make it concrete by demonstrating with real data and proper MATLAB code how the scaling is done. The function ft_transform_vol can be used to scale the volume conduction model. The function ft_electroderealign can be used to position the electrodes on the model. We need a file with 3D electrode locations (in mm), and a matching file with 2D locations for visualisation. To properly document this we also need a volume conduction model (can be a first rough model).


Robert Oostenveld - 2014-06-16 17:20:42 +0200

to do: explain cfg = ... layout = ft_prepare_layout(cfg) and then for i=1... layout.mask{i} = scale*layout.mask{i}; end for i=1... layout.outline{i} = scale*layout.outline{i}; end this keeps the electrodes the same. Scale is a number around 1.00, e.g. 0.95 Show figure before and after using ft_layoutplot.


Robert Oostenveld - 2014-06-27 12:57:17 +0200

Going along the analysis script in detail, it turns out that this is a recurring issue: 1) dealing with different atlas sizes: there is brain atlas and paxinos-franklin. Are they the same size (in mm)? They are certainly not expressed in the same coordinate system. 2) dealing with different size between atlas and actual animal: We use the brainatlas MRI (which comes with a segmented version) to make the volume conduction model. What if the actual animal is not the same size as the animal depicted in brainatlas? 3) dealing with electrodes in the 2D layout: the electrodes are represented in a layout together with an outline of the animal head. The electrodes have a fixed arrangement. How to deal with the layout in case the animal is of a different size? 4) dealing with electrode in the 3D volume conduction model: again the electrodes have a fixed arrangement. The actual animal might have a different size. The placement of the electrodes on the brainatlas head model might not match the placement on the actual animal. 5) dealing with the mapping of source reconstruction results on the actual animal, and dealing with the mapping of source reconstruction results on the atlas: how to map between actual animal, volume model and atlas space?


Jan-Mathijs Schoffelen - 2016-02-22 10:03:55 +0100

What is the status of this? Will it be finalized soon? If not, I suggest to close it for now, because there will probably nobody else who feels strongly enough about this to bring it to a good end.


Robert Oostenveld - 2016-02-22 11:24:45 +0100

agreed, I don't expect much activity here (due to lack of time)


Robert Oostenveld - 2019-08-10 12:32:19 +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 on https://github.com/fieldtrip/fieldtrip/issues.