[FieldTrip] no coil pos for planar combined data

Joram van Driel joramvandriel at gmail.com
Tue Oct 21 21:25:55 CEST 2014


Hi Kaelasha,
I believe (but I might be wrong), that you can use all three sensor types
for Neuromag data (so the uncombined gradiometers *and* the magnetometers)
with lcmv beamforming. That is, the weights (filters) of each voxel are
computed based on covariance between sensors. Even if the different sensor
types catch different aspects of the magnetic field, they probably covary
as a function of, e.g., a stimulus, a cognitive process, or any underlying
brain process. So in that sense, the more data, the better the source
estimate, so why not use all 306 sensors? At least this worked for me and
gave very sensible results.
Maybe other fieldtrip users who use Neuromag data for beamforming can
comment on this?

- Joram

On Tue, Oct 21, 2014 at 6:58 PM, Kaelasha Tyler <ktyler at swin.edu.au> wrote:

> Hi Eelke,
>
> Partly I was just following the tutorials suggestion.
> I am using a Neuromag system. Up until now I have been working primarily
> with the magnetometers.
> I was interested in what difference I would find with data from the planar
> gradiometers.
>
> I had understood that  planar gradiometer sensor pairs measure planar
> magnetic fields perpendicular to each other, at a similar point near the
> scalp. That being the case, I would have thought that using original planar
> gradiometer data in source analysis without combining it would produce in
> nonsensical results. Wouldn't source data at one grid point would then be
> incongruously trying to represent two different sets of information?
>
> Kaelasha
>
> ________________________________________
> From: fieldtrip-bounces at science.ru.nl [fieldtrip-bounces at science.ru.nl]
> on behalf of Eelke Spaak [eelke.spaak at donders.ru.nl]
> Sent: Wednesday, 22 October 2014 2:05 AM
> To: FieldTrip discussion list
> Subject: Re: [FieldTrip] no coil pos for planar combined data
>
> Hi Kaelasha,
>
> Why do you want to do source analysis on combined planar gradiometer
> data? In general, I would say this does not make sense. Instead, I
> would use the original (axial or planar, depending on your system)
> gradiometer data.
>
> Best,
> Eelke
>
> On 21 October 2014 15:57, Kaelasha Tyler <ktyler at swin.edu.au> wrote:
> > Hi all,
> >
> > I am having problems using ft_sourceanalysis with planar gradiometer
> data.
> >
> > I am following a tutorial for beamforming with lcmv method (which I
> believe is a work in progress) found here:
> >
> http://fieldtrip.fcdonders.nl/tutorial/beamformer_lcmv?s[]=covariancewindow
> > Using planar gradiometer data is mentioned in the tutorial, although not
> in detail.
> >
> > My problem:
> > The .avg field produced by time locking, is lost when combining planar
> gradiometers using 'ft_combineplanar'.
> > In this case, I am unable to to use 'ft_sourceanalysis' on combined
> planar gradiometer data.
> > I get the error: Reference to non-existent field 'avg'.
> >
> > I reasoned that I may be able to keep the .avg filed if I combine the
> planar gradiometer data at the outset, prior to timelocking. But in this
> case, the gradiometer positions in data.grad does not include 'coilpos'.
> Eg, no data.grad.coilpos.
> > When calling 'ft_sourceanalysis' in this case, I get the error:
> Reference to non-existent field 'coilpos'.
> >
> > Any suggestions?
> >
> > If the .grad.coilpos was retained when I combined the planar gradiometer
> data at the outset, (after preprocessesing and prior to timelokcing) then I
> would be able to use ft_sourceanalysis on planar gradiometer combined data
> successfully.
> > Is that possible?
> >
> > Thanks in advance! (and thank you to all previous very helpful responses
> on this list).
> >
> > Kaelasha Tyler
> >
> > PhD Candidate
> > Brain and Psychological Science Research Centre
> > Swinburne University of Technology
> > Melbourne
> > Australia
> >
> >
> >
> >
> >
> >
> _______________________________________________
> 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
>



-- 
Joram van Driel, MSc.
PhD student @ University of Amsterdam
Brain & Cognition @ Department of Psychology
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20141021/dbbbb350/attachment-0001.html>


More information about the fieldtrip mailing list