SuperMUDI

The Super-resolution of Multi-Dimensional Diffusion MRI (Super MUDI) dataset contains the data of four healthyhuman subjects with ages range between 19 and 46 years. For each subject 1,344 MRI volumes are provided. Theimaging device was clinical 3T Philips Achieva Scanner (Best, Netherlands) with a 32-channel adult head coil. The Super MUDI Challenge comprises two tasks: isotropic, and anisotropic super-resolution. The names of these tasks were derived from the acquisition strategies of the low-resolution MRI data. The objective of using two down-sampling strategies is to compare the combinations of the down-sampling methods and the super-resolution approaches that can best to be used in a clinical scheme to obtain simulated high-quality and high-fidelity MRI images while reducing the acquisition time. In the anisotropic subsampling the volume has high in-plane resolution (2.5mm ×2.5mm), but thick axial slice (5mm), while in the isotropic subsampling the volume has low resolution (5mm) in all the directions. For our experiments, we use one subject each for training and validation, and two for testing. Reference: Marco Pizzolato, Marco Palombo, Jana Hutter, Vish- wesh Nash, Fan Zhang, and Noemi Gyori, “Super- resolution of Multi Dimensional Diffusion MRI data,” Mar. 2020