Controlling sharpness, SNR and SAR for 3D fast spin echo measurements at 7T by end-to-end learning
Motivation: To non-heuristically identify dedicated variable flip angles (VFAs) simultaneously optimized for PSF and/or SNR of multiple tissues in 3D FSE sequences with very long echo trains at 7T. Goal(s): Flexibly tune between VFAs which favor visibility of small structures (e.g. inner ear imaging), or enhanced signal magnitudes (e.g. low proton-density) Approach: End-to-end optimization which integrates contrast fidelity, SNR and PSF-penalty to minimize image blurring for multiple tissues. Results: PSF-optimization yields significantly reduced image blurring compared to standard T2w/FLAIR vendor sequences with pTx and echo train lengths of 220 while maintaining contrast fidelity. SNR-optimization yields strongly increased magnitudes at equal relative SAR. Impact: This work paves the way to the flexible tuning of the PSF- or SNR-like flavour of dedicated variable flip angles at 7T to enhance visibility of small stuctures or SNR in 3D fast spin-echo sequences with very long echo trains.
Authors
- Patrick Alexander Liebig (ORCID: https://orcid.org/0000-0001-7342-3715)
- Simon Weinmüller (ORCID: https://orcid.org/0009-0002-1163-0772)
- Shaihan Malik (ORCID: https://orcid.org/0000-0001-8925-9032)
- Peter Dawood (ORCID: https://orcid.org/0000-0001-7127-607X)
- Moritz Zaiß (ORCID: https://orcid.org/0000-0001-9780-3616)
- Jürgen Herrler (ORCID: https://orcid.org/0000-0002-4620-8216)
- Peter M. Jakob (ORCID: https://orcid.org/0000-0002-3481-5545)
- Martin Blaimer (ORCID: https://orcid.org/0000-0002-6360-9871)
Publication Details
- Journal
- Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition
- Published
- 2025-09-16
- DOI
- https://doi.org/10.58530/2025/1369
- Primary Topic
- Advanced MRI Techniques and Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.00