Back to the main page.
Bug 2969 - document the use of virtual channels in relation to atlas based ROIs
| Status |
NEW |
| Reported |
2015-09-25 12:09:00 +0200 |
| Modified |
2015-09-25 12:09:04 +0200 |
| Product: |
FieldTrip |
| Component: |
documentation |
| Version: |
unspecified |
| Hardware: |
PC |
| Operating System: |
Mac OS |
| Importance: |
P5 normal |
| Assigned to: |
Hanneke van Dijk |
| URL: |
|
| Tags: |
|
| Depends on: |
|
| Blocks: |
2967 |
| See also: |
|
Robert Oostenveld - 2015-09-25 12:09:04 +0200
There is documentation (meg-uk, chieti, connectivity tutorial) on virtual channel source reconstruction, but all start from a ROI that is defined on the peak in a full-brain source reconstruction. It is also possible to do ROI based analyses starting from an atlas.
Option 1 is to do all in source space and end with ft_sourceparcellate to average over the virtual channel time series.
Option 2 is to use ft_sourceanalysis to compute filters, use ft_sourceparcellate to average the filters and to use ft_apply_montage to estimate the virtual channel time series for the parcels.
The end result is the same, the process is different. Both should be explained. This is to be done by first making a test script that goes through all steps (possibly requiring some improvements to FT code) and subsequently integrating that script with the existing tutorial documentation.
Furthermore, the existing three applicaties of virtual channel source estimates should be updated to use ft_apply_montage (or the cfg.montage option in ft_preprocessing). Of course only after we have confirmed that it works as expected in the test script.