High SNR full brain relaxometry at 7T by accelerated MR-STAT

Area of Science:

• Magnetic Resonance Imaging (MRI)
• High-field MRI Physics
• Quantitative MRI

Background:

• Quantitative relaxometry provides crucial tissue information but is often limited by scan time and signal-to-noise ratio (SNR).
• High-field MRI (7T) offers potential for improved SNR, but requires optimized techniques to manage specific absorption rate (SAR) and radiofrequency (RF) inhomogeneities.

Purpose of the Study:

• To demonstrate the feasibility and robustness of the Magnetic Resonance Spin Tomography in Time-domain (MR-STAT) framework for accelerated, high-SNR relaxometry at 7T.
• To optimize MR-STAT for 7T systems considering SAR constraints and RF inhomogeneities.

Main Methods:

• Developed optimized flip-angle strategies using the BLAKJac-framework with SAR constraints for 7T MR-STAT.
• Mitigated transmit RF inhomogeneities using measured B1+ maps during reconstruction.
• Evaluated sequence robustness and B1+ inhomogeneity sensitivity in phantoms and five volunteers.

Main Results:

• Achieved two-fold acceleration with whole-brain coverage (1x1x3mm³ resolution) in under 3 minutes, including B1+ mapping.
• Demonstrated high accuracy for T1 (1.7% error) and T2 (4.4% error) in phantoms post-B1+ correction.
• In vivo T1 and T2 values in gray and white matter matched literature, with low inter-subject variation (T1: 1.0-2.1%, T2: 4.3-5.3%).
• Observed a 4.1x increase in T1 SNR-efficiency and a 2.3x increase in T2 SNR-efficiency at 7T compared to 3T.

Conclusions:

• Presented an accelerated MR-STAT sequence optimized for 7T MRI.
• Demonstrated high-quality quantitative T1 and T2 parameter maps with significantly increased SNR-efficiency at 7T.
• The optimized MR-STAT framework is robust and feasible for fast, high-quality relaxometry at 7T.