Accelerated T2WI for Liver MRI with Deep Learning Reconstruction: A Prospective Comparison on Image Quality and Respiration Factors

Motivation: T2WI liver MRI can lead to long acquisition times and motion artifacts from respiration. Goal(s): To compare respiratory-gating (RG) and breath-hold (BH) deep learning (DL)-T2WI with traditional T2WI using radial k-space sampling, focusing on the effects of respiratory characteristics on image quality. Approach: A total of 120 participants were prospectively enrolled to undergo all three T2WI sequences. Results: DL-T2WI shows comparable image quality to T2WI based on radial k-space sampling strategies. Variation in abdominal wall position at the end of inhalation affect RG T2WI image quality most. For patients with breath-holding difficulties, RG DL-T2WI may be the preferred alternative. Impact: Respiratory-gating and breath-hold DL-T2WI sequences can be used in daily practice as standard sequences, and parameters derived from the breathing curve can aid in developing a personalized workflow for the imaging process.