Hi Chris et al,
Here is a brief study of the effect of the origin choice with the '-trans' operation. I hope that it is useful, and I am happy to answer any further questions.
BR, Jukka
Jukka Nenonen | Method Development Manager
Elekta Oy
Box 34, FI-00531 Helsinki, Finland
Visiting Address: Siltasaarenkatu 18-20 A, Helsinki
Office: +358 9 756 24085 | Mobile: +358 400 249 557 | Fax: +358 9 756 24011
[log in to unmask]<mailto:[log in to unmask]> | www.elekta.com<http://www.elekta.com>
[cid:imagedf607f.JPG@dafe1340.4f88ebd3]
----------------------------
This message is confidential and may be legally privileged or otherwise protected from disclosure. If you are not the intended recipient, please telephone or email the sender and delete this message and any attachment from your system; you must not copy or disclose the contents of this message or any attachment to any other person.
Any views expressed in this message are those of the individual sender, except where the sender specifies and with authority, states them to be the views of Elekta. Elekta may monitor email traffic data.
----------------------------
________________________________
From: Christopher Bailey <[log in to unmask]>
Sent: 31 January 2018 13:54:53
To: Nenonen, Jukka
Cc: [log in to unmask]
Subject: Re: Maxfilter: running "-trans default" on already SSS'd files
Thanks Jukka,
I’m still a little unsure whether I understand the ‘recomposition’ in my step 1b). Does the file output by maxfilter contain the expansion coefficients estimated on the original input raw data? If not (which I thought was the case), what you’re saying is essentially that in the chain
raw -> basis ‘1a’ -> tsss_mc -> basis ‘1b'
bases ‘1a’ and ‘1b’ are identical (to numerical accuracy)? Ah, perhaps now I understand your comment about the origin too: this only has a chance of being true if the origin for the bases is the same. Makes sense.
Looking forward to learning more about this in the report you mention!
/Chris
On 31 Jan 2018, at 12.35, Nenonen, Jukka <[log in to unmask]<mailto:[log in to unmask]>> wrote:
Moi Chris,
In principle, both ways are fine if the SSS origin is set properly.
If you are doing only source-level analysis, you should not use the trans-option at all and 1a) is the choice.
If sensor-level signals are compared, you can continue to 1b) if 1a) was already done. Then you need to add options ‘-autobad off –force’, because by default there are no bad channels after 1a), and ‘-force’ is needed because otherwise maxfilter complains that SSS was already applied and stops. In 1b, the program recomposes the inside basis amplitudes, and reconstructs the MEG signals corresponding to the default head position.
Using 2), you save time and disk space. However, I recommend to use tSSS always together with movecomp, because head movements often generate nearby artifacts which are not removed by plain SSS.
I am finalizing a report on the effect of the origin choice in trans-operations, and will share it with you soon.
BR, Jukka
Jukka Nenonen | Method Development Manager
Elekta Oy
Box 34, FI-00531 Helsinki, Finland
Visiting Address: Siltasaarenkatu 18-20 A, Helsinki
Office: +358 9 756 24085 | Mobile: +358 400 249 557 | Fax: +358 9 756 24011
[log in to unmask]<mailto:[log in to unmask]> | www.elekta.com<http://www.elekta.com/>
<image31b30d.JPG>
----------------------------
This message is confidential and may be legally privileged or otherwise protected from disclosure. If you are not the intended recipient, please telephone or email the sender and delete this message and any attachment from your system; you must not copy or disclose the contents of this message or any attachment to any other person.
Any views expressed in this message are those of the individual sender, except where the sender specifies and with authority, states them to be the views of Elekta. Elekta may monitor email traffic data.
----------------------------
From: Announcement for the Neuro MEG list [mailto:[log in to unmask]] On Behalf Of Christopher Bailey
Sent: 29. tammikuuta 2018 16:38
To: [log in to unmask]<mailto:[log in to unmask]>
Subject: Maxfilter: running "-trans default" on already SSS'd files
Dear Jukka,
Are these two Maxfilter-pipelines equivalent:
1a) Run maxfilter with tSSS and movecomp, compensate movement to initial head position -> output_1a_tsss_mc.fif
1b) Use output_1a_tsss_mc.fif as input to maxfilter, which is run with -trans default as the only option
2) Run maxfilter with tSSS and movecomp with -trans default as the only option (no temporal extension, plain SSS)
If yes (1 & 2 are equivalent), can you shed any light on how 1b) actually implements the translation? Are the SSS coefficients estimated anew from the now tSSS + MC’d data (output_1a_tsss_mc.fif)? Then I would expect the results to differ, though…?
Best,
Chris
--
Christopher Bailey, MSc
MEG Engineer, MINDLab Core Experimental Facility
Center of Functionally Integrative Neuroscience (CFIN)
Aarhus University, Denmark
tel. cell: +45-2674-9927
tel. office: +45-7846-9936
