Single-Pass Object-Adaptive Data Undersampling and Reconstruction for MRI

There is recent interest in techniques to accelerate the data acquisition process in MRI by acquiring limited measurements. Sophisticated reconstruction algorithms are often deployed to maintain high image quality in such settings. In this work, we propose a data-driven sampler using a convolutional neural network, MNet, to provide object-specific sampling patterns adaptive to each scanned object. The network observes limited low-frequency k-space data for each object and predicts the desired undersampling pattern in one go that achieves high image reconstruction quality. We propose an accompanying alternating-type training framework that efficiently generates training labels for the sampler network and jointly trains an image reconstruction network. Experimental results on the fastMRl knee dataset demonstrate the capability of the proposed learned undersampling network to generate object-specific masks at fourfold and eightfold acceleration that achieve superior image reconstruction performance than several existing schemes.