End-to-end MPI image reconstruction with dual-task generative adversarial network

Traditional reconstruction methods, such as system matrix and x-space, are either extremely time-consuming or result in very blurry images. Here, we propose a novel dual-task generative method to realize high-quality MPI image reconstruction. In this method, the generative model simultaneously undertakes two MPI image processing tasks: reconstruction and segmentation. The main task of image reconstruction generates MPI images, while the auxiliary task of image segmentation guides the main task to focus on key areas of objects in MPI images during the generation process. Our experimental results showed that the proposed dual-task model, with its superior generalization ability, outperforms both traditional MPI reconstruction methods and single-task generative methods. Our results also suggested that the tasks of image generation and image segmentation significantly promote each other during the MPI image reconstruction.