[FieldTrip] Elekta/Neuromag's Maxfilter motor correction

[Gianpaolo Demarchi]

Dear Charly,
On 17. Aug 2021, 18:35 +0200, Billaud, Charly (Research Student) via fieldtrip <fieldtrip at science.ru.nl>, wrote:
> Dear community,
>
> I have been reading about how reducing dimensionality in Maxfilter distorted the phase and may impact posterior resting state connectivity analyses, and wanted to know if it would happen if I only applied the motor correction (using HPIs) without using the SSS/TSSS filters? I struggle to find information about what exactly is done to the data when motor compensation is applied.
In the Elekta/Megin world you can’t. Movement correction (continuous or not) is heavily tied into the Maxfilter/SSS algorithm ...
Back in 2013/14 we tried (and failed) to fit some dipoles at the coils frequencies to get the head position, and correct for it using ft_regressconfound, much like it has been done with the CTF system.
Straight in this way it didn’t work, we had to extract the positions of the coils (“quaternions”),  using Maxfilter (used back then for this only purpose), and then feeding those into ft_regressconfound.
If you search the archive you’ll see that this story periodically resurfaces, but we never had neither the time nor the willingness and man power to bring it to an end ...
Best,
-g

PS: could you send me some references where you were “… reading about how reducing dimensionality in Maxfilter distorted the phase and may impact posterior resting state connectivity analyses”? Tnx ..
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20210817/e0d465ef/attachment.htm>