search

korbinian90 / ROMEO

Executables for ROMEO unwrapping for Linux, Windows and Mac OSX
40 stars 0 forks source link

Question about preprocessing steps for the phase and magnitude data #19

Open ftaheriyan opened 7 months ago

ftaheriyan commented 7 months ago

Which preprocessing steps would you suggest for the phase and magnitude data before using them for the ROMEO?

korbinian90 commented 7 months ago

Usually there is no further preprocessing required when the data comes from the scanner. However, there are some artefacts from other preprocessing steps that ROMEO can not handle.

If the phase is corrupted with open-ended fringe lines, ROMEO cannot unwrap it. I think only Laplacian unwrapping works in that case, but there is usually lowered phase quality around the corrupted regions.

The open-ended fringe-lines occur, when the coil combination is performed with a poor algorithm. For 3T scanners the adaptive combine from Siemens usually works well. For 7T, I would suggest ASPIRE/MCPC3DS, which is also available in mritools.