Mrdegibbs and dwidenoise

#1

Hi all,

I am preprocessing some data that are aquired with an AP phase encode and then a B0 in PA.

I hence would like to run dwibiascorrect as:

dwipreproc DWI_denoise_degibbs.mif DWI_denoise_degibbs_preproc.mif -rpe_pair -pe_dir AP -se_epi AP_PA.mif

Regarding the two functions referred to in the title does it make any sense to run them on the AP_PA.mif image (used by topup) or is this unnecessary?

Claude

#2

I don’t think running dwidenoise on the AP_PA.mif is going to make any difference, there’s not enough redundancy in the data for denoising – and the mismatch between the images due to the different phase-encode directions reduces redundancy further. That said, it’s unlikely to introduce any artefacts, so you could just try it and see what effect it has.

On the other hand, mrdegibbs operates on a single slice at a time, and there’s no reason why it wouldn’t work as expected. It might also be beneficial, and improve the estimation of the B0 field, since Gibbs ringing artefacts can be very prominent in the b=0 images. I haven’t looked into this in detail, but I’d be interested to hear if it does help… :grinning:

#3

Will try both ways and try to report back