Using phase offset

[Ali-Rz]

Hi,

I have a multi-coil low-resolution three-echo GRE field-map data with bipolar readout and a high-resolution single echo GRE data. I wanted to use ROMEO for field-map data to acquire phase offsets for each coil and then coregister to high resolution data and by using complex subtraction, remove the phase offsets from each coil of the high-resolution data to reconstruct qsm maps. My question is, what is the phase offset of each coil in ROMEO output with -q flag? Is it the angle of the complex phase offset?

Thank you in advance for your help.
Ali.

[korbinian90]

Hi Ali,

sounds like an interesting project!
The phase offset calculation is done by MCPC3DS (ASPIRE) and only used inside ROMEO for phase offset correction. There is the flag --write-phase-offsets in ROMEO to write the angle of the phase offset.

To interpolate them to a higher resolution and avoid problems at singularities/wraps you can first convert them to complex, then interpolate the real and imaginary part independently.

Hope that helps!

[Ali-Rz]

Hi Korbinian,

Thank you for detailed explanation.
Actually, I use --write-quality-all to have the offset maps for all the channels, which I assume based on your explanation it saves the angle of the phase offset for each coil in low-resolution data. To convert the phase offsets to a complex value, should I use the magnitude of each coil in exp(1joffset_k) or only exp(1joffset_k) is enough?

Ali.

[korbinian90]

The --write-quality-all writes ROMEO quality maps, they are used to guide the phase unwrapping. They are not related to phase offsets.

It is a good question, if to use the magnitude or not for phase offset interpolation. I think it doesn't make a difference in this case, so I would do it without magnitude.
For phase offset smoothing (for example inside ASPIRE), I preferred to not use the magnitude, as I feared that it could bias the smoothing in the direction from stronger to weaker magnitude.

[Ali-Rz]

Sorry I sent the wrong flag, I use --write-phase-offsets to save the offsets for each channel.
I keep on without magnitude files and keep you posted.

Thank you for your help and detailed explanation.
Ali.