DataProcessing3T

Last-modified: 2023-01-20 (Fri) 01:43:59 (464d)

FolderStructure

###--- Under construction ---###

Individual data processing

>dcms2sdt 0007 <RET> : dicom to sdt data conversion(unifying the multiple dicom files and doing demosaic)
The data type of output sdt/spr files is float.
If you want to transfer the data to the other software,
>sdt2nii m0007 <RET>
Then NIFTI-1 format file is ready as all.nii

For the data overlaying, MRIcro, fsleyes , freeview, etc. can be used.

All data processing

>dcmall <RET>
Process all of the folders(00??) and make sdt/spr files.

EPI data(fMRI data)

Making peak.bit files

car_anlz 170101-01.car +<RET>
Now you get acq.peak.bit, cardio.peak.bit and respir.peak.bit.
(You can still use the old 'peak' command to make the bit files)

Edit the acq.peak.bit file

start from m0007.sdt/spr and peak.bit files
You need to correct the acq.peak.bit file.
>acqbitc acq.peak.bit acq.peak.new.bit 37<RET>
The number, N=37,  need to be adjusted to fit to your data. The acquisition pulses were recorded for individual volumes but we need to have same number of triggers as the number of acquisition(e.g. number of slices(or slice sets for MB) x number of volumes). It must be equal to the number of timing recorded as mosaic_timmng. We'll put the number of additional triggers, so in N=37 case,38 triggers were prepared for each volumes after the acqbitc processing.

Main postprocessing

mcafni m0007<RET> Motion correction by afni. mc2dafni is also available for 2d motion correction.
glmdenoise m0007-aMC +<RET> Remove the spike or the step noise caused by subject's motion
despike m0007-aMC_gdn<RET> Remove the slow drift and sharp spikes
physiofix m0007-aMC_gdnDs acq.peak.new.bit resp.peak.bit card.peak.bit<RET> Retrospective physiological noise removal
ssc2 m0007-aMC_gdnDs-pp<RET> Slice scan time correction
...
(sdtt, sdtcc, sdtglm,...)

Subject motion check

After mcafni, we have *.mc.1D ascii files as results for the motion correction. "mcchk" detects motion error on the files.

>>mcchk 2 2 2 3 3 3<RET> 2deg for rotation, 3 degree for translation (maximum motion size test)
>>mcchk -2 2 -2 2 -2 2 -3 3 -3 3 -3 3<RET> Absolute motion value test. If the motion goes beyond the setting, report it.

DTI data

To make gradient table file,
>dcms2gtb<RET>
Then grd.txt has gradient(x,y,z)  and b-value
You can use this for Diffusion Toolkit(Trackvis).
This procedure is likely unnecessary because Diffusion Toolkit has standard Siemens gradient setting as defaults.
>gtbrot grd.txt d0003.spr<RET>
gtbrot rotates the vectors in the gradient table and make grdR.txt as the output.

Surface rendering of T1W anatomical data with FreeSurfer

>recon-all -i d0010.nii -s subject_ID -all <RET>
>recon-all -i d0010.nii -s subject_ID -all -hires -openmp 12<RET>
To check your processed data, you can do...
>fschk subject_ID<RET>   (Check the volume segmentation results with the rendered surface)
>fschk subject_ID s<RET> (Check the surfaces with the maps  [Alt+c to toggle maps])

DICOM parameter meanings

0051:10160051:1019
AbbreviationImage TypeAbbreviationImage Type
p(x)Grappa acceleration factor xSAT (y)Number of saturation regions
P(x)Sense with acceleration factor xPFPPartial Fourier for Phase(it's PPF in the manual)
NORMNormalization(Prescan image normalization or coil sensitivity normalization)A(y)Number of averages
DIS2D2D distortion correctionIRInversion recovery
DIS3D3D distortion correctionFSfat saturation
NDnon-distortion correction imageWEWater excitation
MRephased image (magnitude image)WSWater saturation
BLACK IMAGEGraphic image or Graphics and table imageDBDark blood
AVERAGEVRT imageADDImage resulting from an addition
ADCApparent Diffusion Coefficient MapSRSaturation recovery
G(x)Grid tag with x mm grid distance
L (x)Line tag with x mm line distance
MTMTC pulse