Distortion Correction of Diffusion-Weighted Image by FSL Learning Model Using 3D U-net [Presidential Award Proceedings]

Abstract

　Pre-processing is essential for various quantitative diffusion analyses. In particular, noise reduction, Gibbs artifact removal, and distortion correction using FSL (TOPUP & EDDY) are performed. However, these pre-processing calculations take a long time. We aimed to simplify the distortion correction technique. A learning model for distortion correction was constructed by 3D U-net using image data before and after FSL distortion correction. Compared to Pre FSL DWI, predicted DWI improved the SSIM/PSNR and exhibited a significant difference. As a result, the learning model of FSL using 3D-U-net was able to correct the distortions in the high b-value DWI.