Accelerated T2* Mapping of the Human Brain at 7T Using Deep Learning: Achieving 0.6 mm Isotropic Resolution in Under 6 Minutes

Motivation: T2* mapping is an important tool for evaluating of healthy and pathological tissues. High-resolution acquisitions can detect submillimeter anomalies but require substantial acceleration for feasible scan times. Goal(s): To demonstrate feasibility of brain T2* mapping at a 0.6mm isotropic resolution in 6min at 7T. Approach: A multi-echo GRE sequence was acquired from four volunteers using incoherent undersampling with an initial acceleration factor of 4. These datasets were then retrospectively undersampled up to a factor of 8 and reconstructed using either conventional or deep learning-based methods. Results: The DL-based reconstruction outperforms conventional methods, enabling acceleration up to 8 with minimal impact on T2* maps. Impact: We demonstrate the efficacy of deep learning-based reconstruction for highly accelerated acquisitions, enabling 0.6mm isotropic R2* mapping of the brain in 6 minutes at 7T. This method highlights submillimeter T2* contrast, potentially enhancing its application in detecting microstructural alterations.