Rapid High-Resolution Prostate Diffusion MRI using Eddy Current-Nulled Convex Optimized Diffusion Encoding and Random Matrix Theory Denoising

Motivation: High-resolution (HR) DWI, although beneficial for prostate tissue characterization and cancer diagnosis, suffers from low signal-to-noise ratio (SNR). The common strategy of averaging to increase SNR prolongs the acquisition time (TA). Goal(s): Decreasing the number of repetitions for averaging while maintaining the SNR for HR prostate DWI. Approach: Eddy Current-Nulled Convex Optimized Diffusion Encoding (ENCODE) combined with random matrix theory (RMT)-based denoising to reduce the repetitions while maintaining SNR. Results: HR-ENCODE-RMT (TA=2 min 30 sec) improved SNR for rapid prostate HR-DWI, and achieved consistent and precise apparent diffusion coefficient (ADC) mapping compared to standard-resolution bipolar DWI (TA=5 min 50 sec). Impact: Eddy Current-Nulled Convex Optimized Diffusion Encoding combined with Random Matrix Theory-based denoising enables rapid high-resolution prostate DWI using fewer repetitions while maintaining the signal-to-noise ratio and robustness of apparent diffusion coefficient maps. This method could improve characterization of prostate cancer.