Back to the main page.
Bug 1339 - wrong number of sensors from bti2grad
| Status |
CLOSED INVALID |
| Reported |
2012-02-23 07:26:00 +0100 |
| Modified |
2012-04-02 16:26:45 +0200 |
| Product: |
FieldTrip |
| Component: |
fileio |
| Version: |
unspecified |
| Hardware: |
PC |
| Operating System: |
Linux |
| Importance: |
P3 normal |
| Assigned to: |
Jan-Mathijs Schoffelen |
| URL: |
|
| Tags: |
|
| Depends on: |
|
| Blocks: |
|
| See also: |
|
Yuval - 2012-02-23 07:26:54 +0100
when using ft_read_header the "gradiometers" for our 4D248 layout are 271
in the grad structure we get 271 labels with 276 points, orientations etc.
this is because bti2grad sums the number of coils and for some gradiometers 2 coils are given per point.
I attached a header from read_4d_hdr.
at the moment I fixed it by adding
numcoils(numcoils>0,1)=1;
just before
totalcoils = sum(numcoils);
Yuval - 2012-02-23 07:29:57 +0100
Created attachment 231
hdr from read_4d_hdr
Jan-Mathijs Schoffelen - 2012-02-23 09:45:19 +0100
Hi Yuval,
This is actually intended. The labels refer to the sensors, the pnt and ori to the coils.
Coils don't have a label. The grad.tra matrix links the labels with the pnt/ori. Each row (corresponding to a sensor) gives the weights with which the individual coils contribute to that sensor. For unbalanced magnetometers there's a one-to-one relation between coils and sensors, a gradiometer requires 2 coils to be properly described. With five gradiometer references (out of 23 reference sensors), there are 5*2 + 18*1 coils. Add this to the 248 magnetometer coils of the true sensors you'll get to 276
Some more information can be found on: http://fieldtrip.fcdonders.nl/faq/how_are_electrodes_magnetometers_or_gradiometers_described
Note that since some time FieldTrip uses a slightly different representation of the gradiometers, making the distinction between sensors and coils more clear. To get to this new representation you can do:
grad = bti2grad(hdr);
gradnew = ft_datatype_sens(grad);
Let me know if things are not clear.
JM