search

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

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

Dynamic range of coilQA 1/g factor map #58

Closed evaalonsoortiz closed 5 months ago

evaalonsoortiz commented 5 months ago

The 1/g factor maps output by the coilQA sequence appear to have values in the range of 700-1000 (see image below).

image

From what I've seen in the literature, 1/g should be in the range of 0-1 (with higher values indicating more noise amplification). Ex, see figure 4: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3131444/

Does anyone know why the dynamic range of our 1/g maps is not between 0-1?

dpapp86 commented 5 months ago

Hello!

The scaling comes directly from the sequence and the DICOM files. As per the WIP guide and the image comments (see https://github.com/spinal-cord-7t/coil-qc-code/issues/57#issuecomment-2089111206), the scaling of the 1/g maps if 0-1000, corresponding to a 1/g of 0 to 1, which in turn corresponds to a g factor of infinite (infinitely bad) to 1 (really good), respectively.

We do not see the full dynamic range used (that is, our range is 1000-700, not 1000-0, because the coilQA never measures a 1/g of less than 0.7, which corresponds to a g-factor of 1.42, which is pretty good. Does the lowest value really correspond to 700-ish? Seems a bit too good to be true.

evaalonsoortiz commented 5 months ago

Ok, so the 1/g should be scaled to 0-1 (divided by 1000)? The range of values was between 0-1000, I was just showing the images between 700-1000, sorry if I wasn't clear.

Here are the re-scaled maps: image

If I understand correctly, this all looks normal (and good for CRMBM and UCL).