Deep-learning-based transformation of magnitude images to synthetic raw data for deep-learning-based image reconstruction

This study demonstrates using generative deep learning for transforming magnitude-only images into synthetic raw data for deep-learning-base image reconstruction. Using relatively few raw datasets, a set of neural networks was trained to generate the missing phase and coil sensitivity information in magnitude-only images. These maps are then recombined into synthetic raw data. We trained end-to-end variational networks for 4-fold accelerated compressed sensing reconstruction on the FastMRI dataset, with increasing training set size. Synthetic raw data showed similar improvements as real raw data with increasing data. This shows promise for applying deep-learning-based image reconstruction when raw data is scarce.