[FieldTrip] Mismatch between channel label and position in hdr.grad (Neuromag 306 system)

[Jarang Hahm]

Dear Hanneke,


Thanks to you, I properly found neighbours. :-)

And I confirmed it by plotting the neighbours. To do this, I modified a
code 'ft_plotneighbours' for correcting a mismatch between the channel
position and label.

You also probably need this code, so I have attatched what I modified. :-)


Thanks!


Best,


Jarang Hahm


On Mon, Jun 25, 2012 at 6:01 PM, Hanneke van Dijk <
Hanneke.vanDijk at med.uni-duesseldorf.de> wrote:

> Dear Jarang,
>
> Coincidentily I have been working on the same problem a few weeks ago.
>
> First of all you should take care that you repair a horizontal gradiometer
> with neighbouring horizontal gradiometers and vertical ones with vertical
> ones. This is not built into fieldtrip yet.
>
> I have have made a partial fix, which works with the 'layout' option but
> not with the 'grad' method. I have attached the function I addapted from
> Fieldtrip, and the layout files that seperatly name the different
> gradiometers (made by Elisabeth May ;-) )
>
> I hope it somehow helps you!
>
> Best regards,
>
> Hanneke van Dijk
>
> ________________________________________________________________
>
>
> Hanneke van Dijk, PhD
>
>
> http://www.uniklinik-duesseldorf.de/deutsch/unternehmen/institute/KlinNeurowiss/Team/HannekevanDijk/page.html
>
> Institute for Clinical Neuroscience,
>
> Heinrich Heine Universität Düsseldorf, Germany
>
> Hanneke.vanDijk at med.uni-duesseldorf.de
>
> Tel. +49 (0) 211 81 13074
>
> __________________________________________
>
>
> On Mon, Jun 25, 2012 at 4:57 AM, Jarang Hahm <jarang.hahm at gmail.com>wrote:
>
>> Dear FieldTrip user.
>>
>>
>>
>> Hi, I'm processing MEG data of Neuromag 306 system after removing some
>> bad channels using fieldtrip-20120620 toolbox.
>>
>>
>>
>> To repair the bad channels with its neighbours, I firstly found the
>> neighbor of all channel based on distance using 'ft_prepare_neighbours'.
>>
>> However, the outcome of neighbour was far more than default distance,
>> 4cm, from a target channel.
>>
>> So, I checked thoroughly the code 'ft_prepare_neighbours', finding it
>> used information of channel label and position from data’s hdr.grad to find
>> neighbour of the target.
>>
>>
>>
>> The problem is channel label and position in hdr.grad are arranged
>> differently.
>>
>> As far as I found, both information come from different source: the
>> channel label was from code 'fiff_read_meas_info', while the channel
>> position was from ‘channelposition’ in 'ft_datatype_sens'.
>>
>> And  the channel label is sorted into the sensor location, e. g. MEG
>> 0113, MEG 0112, MEG 0111, MEG 0122, etc, however, the channel position is
>> arranged along the sensor type (magnetometer and gradiometer) according to
>> its ending number (1, 2, or 3), e.g. MEG 0111, MEG 0121, MEG 0131, ....,
>> MEG 0112, MEG 0122, MEG 0132, ...MEG 0113, MEG 0123, MEG 0133, etc.
>>
>> Thus, the label and position of channel in hdr.grad were mismatched so
>> that we got somewhat weired neighbour after using 'ft_prepare_neighbours'.
>>
>>
>> It should be matched correctly to find the proper neighbour and ensure
>> the result of repair as well.
>>
>> In fact, the code 'channelposition' in 'ft_datatype_sens' yield the
>> channel label as well as the channel position. It might help to solve this
>> problem.
>>
>>
>>
>> Is there any other way to solve this problem?
>>
>>
>>
>>
>> Best,
>>
>>
>> Jarang Hahm
>>
>>
>>
>> _______________________________________________
>> fieldtrip mailing list
>> fieldtrip at donders.ru.nl
>> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>>
>
>
> _______________________________________________
> fieldtrip mailing list
> fieldtrip at donders.ru.nl
> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>



-- 
함자랑 드림.

Jarang Hahm

Interdisciplinary Program in Cognitive Science
Seoul National University
Seoul, Korea, Republic of.
+82-10-9079-3121
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20120627/a52699f0/attachment-0002.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: neuromag_neighbourplot.m
Type: application/octet-stream
Size: 7578 bytes
Desc: not available
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20120627/a52699f0/attachment-0002.obj>