Sharing

revisiting for ENIGMA MEG share. (2023-01-18)

Acquisiton

Instructions documented in a word document

Processing?

see ref_scripts/.

Highlights from /data/Luna1/Multimodal/Rest_2015_97subjects/scripts/ss_maxfilter.sh

echo "TRANS"
/neuro/bin/util/i686-pc-linux-gnu/maxfilter-2.2 \
	-f /Volumes/Zeus/meg/MMY4/subjs/${ss}/${ss}_Switch_run${i}_sss.fif \
	-o /Volumes/Zeus/meg/MMY4/subjs/${ss}/${ss}_Switch_run${i}_ds_sss.fif \
   -origin fit -trans default -frame head -force -v -autobad off -ds 4 \
    >> /Volumes/Zeus/meg/MMY4/subjs/${ss}/${ss}_Switch_run${i}_trans.log

And maxfilter_rest_wrapper.sh

RESTDIR="/data/Luna1/Multimodal/Rest_2015_97subjects/subjects"
RAWDIR="/data/Luna1/Multimodal/MEG_Raw"
/neuro/bin/util/i686-pc-linux-gnu/maxfilter-2.2 \
	-f $RESTDIR/${ss}/MEG/${ss}_emptyroom_raw_chpi_sss.fif \
	-o $RESTDIR/${ss}/MEG/${ss}_emptyroom_ds_sss_raw.fif \
	-origin fit -trans default -frame head -force -v -autobad off -ds 4 \
	> $RESTDIR/${ss}/MEG/${ss}_emptyroom_trans.log

Transformation

fif file header

(20240327) See 00_findtrans.bash (using find_trans.py; makes trans/files_luna1MM.txt and trans/log_rest.txt).

Writes file per rows with columns like file, hit/partial/miss, from, to, mean(trans), trans_matrix

where "hit" requires from=head, to=MRI, and mean(trans)!=1. "partial" given when to and from match but trans mat mean is 1. "miss" is all other files (from and to not as wanted).

/Volumes/Medusa/Luna1/MM/FS_Subjects/10923_20120119/mri/T1-neuromag/sets/COR-hwangk-120326-110007.fif	hit	head	MRI (surface RAS)	3.9254002512898296 0.9980863332748413 -0.04823729395866394 0.03868649899959564 -0.003494652220979333 0.054638467729091644 0.9809278249740601 -0.1865348219871521 -0.006118252873420715 -0.028950825333595276 0.1882915496826172 0.9816863536834717 -0.043580930680036545 0.0 0.0 0.0 1.0

No from head to MR for rest?

trans log for MMY4 rest saves to rest_ds_ss.fif

tail -n1 /Volumes/Zeus/meg/MMY4_rest/subjs/11403/11403_Switch_rest_trans.log
   EXIT OK: Wrote successfully the result to file ... 11403_Switch_rest_ds_sss.fif.

but header there is from MEG deviced to head

grep 11403_Switch_rest_sss.fif trans/files*
...11403_Switch_rest_sss.fif       MEG device      head    4.049433422274888 0.9982056021690369 ...

Many COR files?

ls /Volumes/Medusa/Luna1/MM/FS_Subjects/11059_20130401/mri/T1-neuromag/sets
COR.fif                      COR-scott-150616-161439.fif
COR-scott-150616-160940.fif  COR-scott-150616-161528.fif
COR-scott-150616-161036.fif  COR-scott-150616-161610.fif
COR-scott-150616-161127.fif  COR-scott-150616-161702.fif
COR-scott-150616-161211.fif  COR-scott-150616-161751.fif
COR-scott-150616-161305.fif  COR-scott-150616-161843.fif
COR-scott-150616-161352.fif  COR-scott-150616-161940.fif

COR.fif has identity trans matrix. No idea what the other files are. all have file size 6.6Mb.

log file

Separately, looking at transformation annotated in log file. These go from device to head (not what we want?)

Like

10637_rest_trans	  0.9999  -0.0045   0.0102 !  1.0000   0.0000   0.0000
10637_rest_trans	  0.0047   0.9997  -0.0246 !  0.0000   1.0000   0.0000
10637_rest_trans	 -0.0101   0.0246   0.9996 !  0.0000   0.0000   1.0000
10637_rest_trans	 -0.4250  -2.4821  57.0022 !  1.8484  10.5242  40.2005
10637_rest_trans	Rotation change -0.0, -0.0, -0.0 degrees
10637_rest_trans	Position change (-2.3, -13.0, 16.8) = 21.4 mm

Written output fif file is miss w/ find trans:

grep -i 'EXIT.*.fif' 10637_rest_trans.log # .../MEG/10637_rest_ds_sss_raw.fif

(from MEG device to head)

10637_rest_ds_sss_raw.fif   miss    MEG device      head    4.052573025692254 1.0 0.0 0.0 0.001848407555371523 0.0 1.0 0.0 0.01052415743470192 0.0 0.0 1.0 0.04020046070218086 0.0 0.0 0.0 1.0