Sonic DLTM Enhanced 3D Double-Echo Steady-State Sequence for Improved Imaging Resolution in Magnetic Resonance Neurography

Motivation: High spatial resolution is crucial for MR neurography to accurately depict small peripheral nerves and associated pathology. However, achieving high-resolution images is often constrained by scan time and signal-to-noise ratio (SNR). Goal(s): This study explores the feasibility of using a deep learning acquisition and reconstruction method for highly accelerated imaging (Sonic DLTM) to improve imaging resolution in a 3D double-echo steady-state sequence (MENSA NERVE). Approach: The study compares images from reference and proposed sequences in patients undergoing routine upper extremity and neck MR neurography. Results: Preliminary results indicate that the prototype Sonic DLTM can achieve higher imaging resolution at the same scan time. Impact: The higher acceleration factors enabled by Sonic DLTM can be utilized to achieve higher spatial resolution in MENSA NERVE, leading to better nerve morphology depiction.