AW: [FIELDTRIP] ft_channelrepair with neuromag data

Tue Oct 26 10:46:46 CEST 2010

Dear Nina,

indeed this looks funny. I have seen that maxfilter sometimes has 
produced unexpected results, but usually the channels have less noise 
after transformation. Your displays demonstrate the problem, but provide 
insufficient information to search for origin of it. I think, it is 
necessary to place one of these files to an ftp-server together with the 
list of bad channels you have used and the version of maxfilter. I would 
recommend to ask Jukka Nennonen for help because he is the maxfilter-pro 
and the displayed result should not show up. We shall continue the 
discussion here when we know the reason for this strange behavior.

best wishes,
Burkhard

Nina Kahlbrock wrote:
> Dear Burkhard,
> I am also working on NM306 data and I have a specific problem repairing
> broken channels using maxfilter. This email might thus be more suited for
> the neuromeg discussion list. However, I thought it might be of interest to
> other NM users as well...
> What I would like to do is pretty much the same as described in the previous
> mails: interpolate broken channels and use a standard head position for
> multiple recordings in one subject.
> I used fieldtrip's automatic and semi-automatic artifact rejection routines
> to identify channels that are flat, show jumps and are noisy. I defined
> these as bad in maxfilter. Then I used sss/tsss to achieve the desired
> steps. It seemed very straight forward. 
> However, when looking at the results, the signal seems to be increased (and
> more noisy and jumpy) in certain channels (see attached pdfs of bad and
> non-bad channel (1st page: raw data, 2nd page: sss data, 3rd page: tsss
> data). Have you ever experienced anything like this? The only reason I can
> come up with is that in maxfilter I used the continuous raw data file and
> not as in fieldtrip, only trials. Between trials there are pauses, where the
> channels could have been affected by a few movements. Could this explain my
> difficulties?
> Is there a way to 'only' interpolate channels and realign the head, without
> impacting other non-broken channels?
> Thank you very much for your answer!
> Nina   
>
>
>
> -----Ursprüngliche Nachricht-----
> Von: FieldTrip discussion list [mailto:FIELDTRIP at NIC.SURFNET.NL] Im Auftrag
> von Burkhard Maess
> Gesendet: Freitag, 22. Oktober 2010 09:25
> An: FIELDTRIP at NIC.SURFNET.NL
> Betreff: Re: [FIELDTRIP] ft_channelrepair with neuromag data
>
> Dear David,
>
> I agree with Jan-Mathijs.
>
> variant a: apply maxfilter - that will repair all broken channels. You 
> also can manually define certain channels as broken. Maxfilter may also 
> align data of different headpositions to a single position.
>
> variant b: use megrealign - it includes a minimum norm localization and 
> from there MEG data is forward computed. In this way, you can repair 
> channels, align different head positions to the same and even convert 
> data between different systems.
>
> all the best,
> Burkhard
>
>
> jan-mathijs schoffelen wrote:
>   
>> Dear David,
>>
>> This is a good point. As far as I can see from the code, no 
>> distinction is made between gradiometers and magnetometers.
>> Naively, I would propose the following: first indeed split the data 
>> into two subset: gradiometers only and magnetometers only, only then 
>> apply the channel reparation routine. Yet, for the gradiometer subset 
>> data, it is questionable whether as such the replacement of a bad 
>> channel with its neighbours makes sense, because this would lead to 
>> averaging gradients of the magnetic field with an orthogonal orientation.
>>  
>> In other words, this is probably not the most sensible approach. I 
>> don't think repairing after combining the planar gradients is 
>>  desired, because this leads to an unwanted amplification of the noise 
>> (because the combination step results in taking an absolute value). 
>> Alternatively, you could consider using ft_megrealign, which in 
>> principle could be used to repair bad channels. The recipe would be to 
>> only use the data with the clean channels in the input, and you would 
>> want to interpolate the data back onto the original sensor-array. An 
>> example for interpolating between two types of CTF-systems is shown 
>> here http://fieldtrip.fcdonders.nl/example/megrealign.
>>
>> Just out of curiosity: doesn't the maxfilter in the Neuromag software 
>> allow for similar things?
>>
>> Best wishes,
>>
>> Jan-Mathijs 
>>
>> On Oct 21, 2010, at 9:49 PM, David Ziegler wrote:
>>
>>     
>>> Hi Fieldtrippers,
>>>
>>> I would like to use ft_channelrepair to interpolate data for some 
>>> missing/deleted channels with Neuromag 306 data.  I am just wondering 
>>> whether there are any potential problems with using the nearest 
>>> neighbor interpolation, given the triplet format of the Neuromag 
>>> system (two planar gradiometers and one magnetometer).  
>>>
>>> If this is problematic, are there other options to fixing bad/missing 
>>> channels (e.g., can ft_channelrepair be run on just a subset of the 
>>> channels, say the gradiometers after I run ft_combineplanar)?
>>>
>>> Thanks!
>>> David
>>>
>>> -- 
>>> David A. Ziegler
>>> Department of Brain and Cognitive Sciences 
>>> Massachusetts Institute of Technology 
>>> 43 Vassar St,  46-5121 
>>> Cambridge, MA  02139
>>> Tel: 617-258-0765
>>> Fax: 617-253-1504
>>> daz at mit.edu <mailto:daz at mit.edu>
>>>
>>>  
>>>
>>>  
>>>
>>>
>>>       
> --------------------------------------------------------------------------- 
>   
>>> You are receiving this message because you are subscribed to 
>>> the FieldTrip list. The aim of this list is to facilitate the discussion 
>>> between users of the FieldTrip toolbox, to share experiences 
>>> and to discuss new ideas for MEG and EEG analysis. 
>>> See also http://listserv.surfnet.nl/archives/fieldtrip.html 
>>> and http://www.ru.nl/neuroimaging/fieldtrip.
>>>
>>>       
> ---------------------------------------------------------------------------
>   
>> Dr. J.M. (Jan-Mathijs) Schoffelen 
>> Donders Institute for Brain, Cognition and Behaviour, 
>> Centre for Cognitive Neuroimaging,
>> Radboud University Nijmegen, The Netherlands
>> J.Schoffelen at donders.ru.nl <mailto:J.Schoffelen at donders.ru.nl>
>> Telephone: 0031-24-3614793
>>
>>
>>     
> --------------------------------------------------------------------------- 
>   
>> You are receiving this message because you are subscribed to
>> the FieldTrip list. The aim of this list is to facilitate the discussion
>> between users of the FieldTrip toolbox, to share experiences
>> and to discuss new ideas for MEG and EEG analysis.
>> See also http://listserv.surfnet.nl/archives/fieldtrip.html
>> and http://www.ru.nl/neuroimaging/fieldtrip.
>>
>>     
> ---------------------------------------------------------------------------
>   
>
>   

-- 
------------------------------------------------------------
Dr. Burkhard Maess
Max Planck Institute for Human Cognitive and Brain Sciences
Stephanstr. 1a, P.O. Box 500355, D-04303 Leipzig
Aussenstelle Bennewitz, phone/fax: +49(3425)887525-26/-11   
mail: maess 'at' cbs.mpg.de,           http://www.cbs.mpg.de
------------------------------------------------------------
Please do not attach files which may contain executable code, 
e.g. .rtf is preferred over .doc(x)

---------------------------------------------------------------------------
You are receiving this message because you are subscribed to
the  FieldTrip list. The aim of this list is to facilitate the discussion
between  users of the FieldTrip  toolbox, to share experiences
and to discuss  new ideas for MEG and EEG analysis.
See also http://listserv.surfnet.nl/archives/fieldtrip.html
and http://www.ru.nl/neuroimaging/fieldtrip.
---------------------------------------------------------------------------