Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

MT protocol #12

Open
mjt320 opened this issue Aug 29, 2024 · 2 comments
Open

MT protocol #12

mjt320 opened this issue Aug 29, 2024 · 2 comments

Comments

@mjt320
Copy link

mjt320 commented Aug 29, 2024

Hello,

Thanks for sharing these protocols! I'm implementing the Siemens version and have a comment/query about the MT GRE acquisitions.

The original MTSat paper by Helms assumes non-selective excitation, such that the MT and excitation pulse flip angles are both proportional to B1+. As a result, there is (some) cancellation of B1+ inhomogeneity effects. The generic protocol uses slab-selective excitation, so the spatial flip angle profiles of the two pulses are different. I wondered if you'd looked at the implication for accuracy of MTSat measurements?

Thanks,
Michael

@jcohenadad
Copy link
Member

Hi Michael,

Great question! It's been many years since the genesis of this protocol, but if I remember correctly the reason we went for the slab-selective was to reduce the amount of artifacts (I forgot exactly what kind of artifacts we were trying to get rid of-- possibly S-I wrapping).

We have not looked at the implication for accuracy of MTSat measurement, but if you'd be willing to acquire with/without slab-selective and compare, that would be awesome. We could assist with the analysis script.

Cheers

@mjt320
Copy link
Author

mjt320 commented Sep 2, 2024

Hello,

Thanks for replying so quickly!
Yes, I agree it makes sense to use slab-selective to avoid wrap without needing masses of oversampling. We'd like to scan all/most of the spine so we might need to change the orientation anyway. We'll try it with/without slab-selective if possible. I'll let you know if we do.

Cheers,
Michael

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants