Hi,
I think the webpage refers to "images affected by large artifacts" in terms of volumes. Here you'd have to throw out the entire volumes and adjust your bvecs/bvals files accordingly (i.e by throwing out the respective entry). You may also eliminate a particular slice from ALL volumes if that slice is affected across the entire series. That's perfectly fine for dtifit (or even bedpost(x)) and you would not have to change your bvecs/bvals in that case. However, it may cause issues for probtrack - not computationally but when it comes to the tracking results and what you are going to do with these. Also you may want to be careful about eliminating slices before or after eddy_correct.
Hope that helps,
Cheers-
Andreas
________________________________
Von: FSL - FMRIB's Software Library im Auftrag von Jerry Chen
Gesendet: Di 28.10.2008 20:27
An: [log in to unmask]
Betreff: [FSL] dtifit and missing data
Hello,
From the processing pipeline, http://www.fmrib.ox.ac.uk/fsl/fdt/fdt_pipeline.html, it is suggested that we remove bad slices with artifacts.
I am using DTI-Studio to do this. However, I would like to know whether or not FSL's dtifit can process data with missing slice data. I have heard that it can not, but I can not find verify this through the website or support e-mail archive.
Thanks for your time,
- Jerry
|