spinal-cord-7t / coil-qc-code

7T Spinal Cord Coil QC Analysis Code
0 stars 0 forks source link

lack of documentation on the coilQA sequence #57

Open evaalonsoortiz opened 2 months ago

evaalonsoortiz commented 2 months ago

The coilQA sequence is available as a C2P from the Martinos Center: https://www.nmr.mgh.harvard.edu/c2p or as a product sequence on the Siemens Terra. I have not been able to find any documentation explaining what the different images are.

@dpapp86 created a post on the Magnetom forum which conveniently summarizes the different images that are output by this sequence.

In the THS_CoilQA_code.ipynb notebook that @dpapp86 prepared (https://github.com/spinal-cord-7t/coil-qc-code/blob/main/THS_CoilQA_code.ipynb), the following comment can be found:

['R 2','R 3', 'R 4', 'R 6', 'R 8', 'R 2 x 2', 'R 2 x 3', 'R 3 x 2', 'R 3 x 3', 'R 3 x 4', 'R 4 x 3', 'R 4 x 4']

which presumably lists all of the 1/g factor maps that are output by the coilQA sequence and what acceleration they correspond to. My question is, where did @dpapp86 get this information? Is there any available documentation on the coilQA out there?

dpapp86 commented 2 months ago

Hello!

1, I do not have any published referencable description of the sequence. 2, However, I do have the WIP documentation of the old CoilQA package, I have sent it to your email adress. 3, The R factors can be seen in the ImageComment tag of the DICOM outputs. However, dcm2niix crushes all of those comments into the comment equivalent with the first one. That is, the DICOM files have ImageComments to the effect of "1/G, R=2 [avg=944, Gmax=1.95, scale=1000.000, offset=0.000]", with R=2 to R=4x4, but the dcm2niix conversion only results in the first one. In brief, this is because they are considered as one series by dcm2niix, so the other comments do not get written into the nifti/json pair.

evaalonsoortiz commented 2 months ago

Thank you!

evaalonsoortiz commented 2 months ago

@Kyota-exe for your info, below is a screenshot with the relevant info contained in the document Daniel just sent me. Please confirm that the R = 2 x 2 image with think we are analyzing is indeed the correct one (since this can not be reliably verified from the json files I want to be sure).

Screen Shot 2024-05-02 at 9 04 17 AM
evaalonsoortiz commented 2 months ago

@Kyota-exe Never mind, I just realized the 1/g-factor map we include in the dataset has all of the R maps in the 4th dimension.