End-to-end learning of variable flip angle schemes for sharper 3D MRI of biological samples at 11.7T

Motivation: Achieving high-quality 3D fast spin echo imaging for biological samples is challenging due to T2-dependent blurring and SAR limitations in ultrahigh field MRI. Goal(s): Tailoring variable flip angle (VFA) schemes that balance signal-to-noise ratio (SNR) and point spread function (PSF) characteristics at 11.7T. Approach: Using an optimization framework with SNR- and PSF cost function components, dedicated VFA schemes were computed and tested across a range of SNR and PSF weightings. Results: SNR-optimized VFA schemes yield strongly enhanced signal magnitudes, while PSF-optimized enhance image sharpness, with marked improvement over conventional FSE, demonstrating adaptability for pre-clinical high-field MRI applications. Impact: This work showcases the benefit of dedicated VFA optimization for pre-clinical ultrahigh field MRI by enabling tailored imaging protocols enhancing SNR or sharpness. Its adaptability holds promise for fast and high throughput imaging of biological samples and plant research applications.