Hi Gabriel - The standard pre-processing in tracula is likely to work for all these cases. To make sure that the standard registration-based eddy-current correction worked for all cases, you should look at the final correct images (dmri/dwi.nii.gz) and make sure that all images in the series are well aligned to the b=0 images. Check that this worked as well for the b=1500 protocol as for the lower b protocols.
As part of the pre-processing in tracula, the b=0 images are registered to the subject's T1 image. The T1 image has to be processed with freesurfer to segment various cortical and subcortical structures, and tracula use this segmentation to guide the tractography in the diffusion data. So a T1 image that can be processed with freesurfer ir required by tracula.
Hope this helps, a.y
On Tue, 23 Sep 2014, Gabriel Gonzalez Escamilla wrote:
Dear all,
I have a question regarding the pre-prossesing steps of DTI data. I have 5 different datasets, each with different acquisition parameters, including the number of directions (16, 32 and 64), the b-value (1000, 1300 and 1500) and the magnetic field (1.5 and 3T). Do you think that a single pre-processing procedure can be applied to all of them? or it will vary depending on the acquisition parameters?
A second question would be, if the pre-processing steps of FS include some distortion correction using a high-resolution image? If so, could it be replaced by the T2 or a b0?
Can you point me to the correct literature of this?
Many thanks in advanced for your time and help, Gabriel