Reducing Geometric Distortion in Prostate Diffusion MRI using Reduced Field-of-View ENCODE

Motivation: Clinical diffusion-weighted MRI (DWI), although essential for prostate cancer (PCa) diagnosis, can suffer from geometric distortion. Reduced field-of-view (rFOV) eddy current-nulled convex optimized diffusion encoding (ENCODE) can reduce susceptibility and eddy current-induced distortion. Goal(s): Evaluate the performance of rFOV-ENCODE DWI vs. clinical bipolar DWI in subjects with suspicion of PCa Approach: Quantitative analysis of prostate geometry on DWI using Dice's similarity coefficient (DSC) and Hausdorff's distance (HD) vs. T2-weighted MRI in 60 subjects. Results: rFOV-ENCODE had significantly improved HD and similar/higher DSC than bipolar-DWI, indicating reduced geometric distortion. The most prominent improvements were observed in cases with rectal gas. Impact: The rFOV-ENCODE sequence, combining rFOV acquisition and ENCODE, can reduce geometric distortion for prostate DWI without sacrificing scan time across different scanner platforms. The technique can potentially improve diagnosis of prostate cancer.