Question

Question Posted 11/19/20:
Hi, I am performing EPI correction for DTI data. Since only PA scan was provided and no reverse phase scan was given for EPI correction. I was looking for another method to do the fsl topup processing using B0.nii. Do you know where I can download B0.nii?
Response posted 11/19/20 by Rob Reid:
Hi,

Could you provide some more information about your plan? Did you find a method, and if so what is it? And does it use a field map (often called B0map) or a b=0 image? Remember that B (or B0) is the main field of the scanner and b is the diffusion weighting. b=0 images are available from the diffusion scans themselves, while field maps are separate scans. Using field maps with ADNI is tricky because you need to know several scan parameters, such as echo spacing, that are not always in the DICOM headers, account for head motion, and unfortunately the head coverage in the field maps is often insufficient.

I've had better luck removing EPI distortion by synthesizing a T1 from the diffusion data and using it to calculate the warp (see Bhushan, C. et al. Co-registration and distortion correction of diffusion and anatomical images based on inverse contrast normalization. NeuroImage doi:10.1016/j.neuroimage.2015.03.050.) Apparently some people do the reverse and synthesize a 0 TE b=0 from the T1, and use that as input for topup/eddy (https://www.biorxiv.org/content/10.1101/2020.09.14.260240v2.full )

I hope this helps
Response posted 11/19/20 by qi zeng:
Hi, I used the Synb0-DISCO 1 approach and publication (see. Schilling, K. G. 2020. Distortion correction of diffusion-weighted MRI without reverse phase-encoding scans or field-maps). So I need acqparams.txt, T1.nii and b0.nii(non-diffusion weighted images). You mentioned b=0 images are available from the diffusion scan themselves. Is the code: fslroi dti.nii.gz notif 0 1? Aslo do I download acqparams.txt or I created one?
Thank you so much!
Response posted 11/19/20 by Rob Reid:
The code will be something like that, although I would give the output file something more descriptive, like b0_EPI.nii. You will have to make acqparams.txt yourself based on the instructions in the FSL wiki. (Although apparently the latest eddy can do it for you using the BIDS json file from dcm2niix - I have not tried it yet myself, but it sounds very cool.)

Providing support for FSL, Synb0-DISCO, etc., is beyond the scope of ADNI, but you appear to be on the right path and should figure it out with a little experimentation. Good luck.
Go back to list of topics >
2024 Alzheimer’s Disease Neuroimaging Initiative
This website is funded by the Alzheimer’s Disease Neuroimaging Initiative