Accelerated FLAIR imaging at 0.6T using T2W-guided multi-contrast deep learning-based reconstruction using a Zero-Shot approach

Motivation: Mid-field MRI (0.6T) is an accessible alternative for clinical brain imaging; however, FLAIR images, essential for identifying lesions and edema, often exhibit reduced quality at this field strength. Goal(s): We aimed to improve the accelerated FLAIR scans at 0.6T with deep learning while addressing the challenge of limited 0.6T datasets. Approach: A two-contrast content/style model was trained on an unpaired 3T image dataset and was applied zero-shot to reconstruct FLAIR images with T2W guidance at 0.6T. Results: The proposed method outperformed standard compressed sensing, demonstrating its potential for enhancing mid-field FLAIR image quality for further clinical applications. Impact: We showed that FLAIR and T2-weighted scans can be used for contrast/style-based reconstruction methods, even when trained at 3T data directly applied to 0.6T data. Resulting improved image quality improves the usability of mid-field MRI.