Correct Pipeline #36
Description
Hi,
first of all thank u fir this wonderful software.
I am trying to use it in a study about MS doing some segmentations in different patients with different timepoints to see if the lesion' load in time is the same or not, but I am having some trouble to have consistent results.
My pipeline for FLAIR and MPRAGE is this one:
- Dicom to NIFTI
2)RobustFOV - Bias field correction and Normalisation (between the two different timepoints)
- FLIRT 6DOF between flair and mprage
4)skull stripping
And I use the docker with the -stripped option.
I have like 8-10% of lesion volume variation between 2 timepoints even if to me the FLAIR are essentially the same.ù
Do you have any suggestions or Pipeline to be more consistent (we don't have a Nvidia card at the moment).
thank you,
Roberto