We evaluated images from undersampled data using a U-Net with common metrics (SSIM and NRMSE) and with a model for human observer detection, the sparse difference-of-Gaussians (S-DOG). We also studied how the results vary when changing the loss function and training set size. We saw that the S-DOG model would choose an undersampling of 2X while SSIM and NRMSE would choose 3X. In previous work, human observers also chose a 2X acceleration. The S-DOG model led to the same conclusion as the human observers. This result was consistent with changes in training set size and loss function.
This abstract and the presentation materials are available to members only; a login is required.