Neuromag 306 channel data import

[Lauri Parkkonen]

Hi Yisheng and Robert,

>>> Any one can provide a "cookbook" reference for how to import the
>>> raw and averaged fif files into FieldTrip?
>>
>> Neuromag fif files should be directly accessible by the standard
>> FieldTrip functions for opening/reading MEG/EEG data files. Try
>> first with read_fcdc_header() to get the "meta information" and  then
>> with read_fcdc_data() to actually read some data from the  file.
>> These functions should work both with raw/continuous and  averaged data.
>
> The functions from the meg-pd are not included in Fieldtrip, you have
> to add them seperately to your matlab path.

Yes, sorry, I forgot to mention this.

>> NB: I just downloaded the newest daily FieldTrip release and  noticed
>> that read_fcdc_data is unable to retrieve data when the  requested
>> span extends over a buffer boundary. I will look into  this and try
>> to fix it.
>
>
> I did not know whether the blocks in the data are trials (with a
> potential gap in between) or buffers. In case of continuous blocked
> data, you would set the last argment of
>   dat = read_fcdc_data(datfile, hdr, begsample, endsample, chanindx,
> continuous)
> i.e. the continuous to 1. Then it will not perform the boundary
> check. To get this behaviour in fieldtrip prepropcessing, you should
> specify cfg.datatype='continuous'.

The Neuromag fif files do not support the epoch/trial mode. Thus, the
continuous flag should (always) be set to 1 when working with Neuromag
raw data. However, the raw data files may still contain multiple
segments if the recording was paused, for example, but these segments
are usually much longer than one trial. I had modified my own version of
the read routine in such a way that it assumes continuous files but does
check for possible gaps within the requested span and returns with an
error if there is one. Hence the difference w.r.t. the official version.

To avoid spamming everybody on the list, I'll check with Robert directly
if we can update the Neuromag import routines.

Cheers,
Lauri

--
-----------------------------------------------
Lauri Parkkonen
Brain Research Unit, Low Temperature Laboratory
Helsinki University of Technology
Otakaari 3 A, 02150 ESPOO, FINLAND
tel: +358-9-4512965
fax: +358-9-4512969
mailto:lauri at neuro.hut.fi
http://boojum.hut.fi