English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Magnetic resonance image (MRI) reconstruction is a severely ill-posed linearinverse task demanding time and resource intensive computations that cansubstantially trade off {\it accuracy} for {\it speed} in real-time imaging. Inaddition, state-of-the-art compressed sensing (CS) analytics are not cognizantof the image {\it diagnostic quality}. To cope with these challenges we putforth a novel CS framework that permeates benefits from generative adversarialnetworks (GAN) to train a (low-dimensional) manifold of diagnostic-quality MRimages from historical patients. Leveraging a mixture of least-squares (LS)GANs and pixel-wise $\ell_1$ cost, a deep residual network with skipconnections is trained as the generator that learns to remove the {\italiasing} artifacts by projecting onto the manifold. LSGAN learns the texturedetails, while $\ell_1$ controls the high-frequency noise. A multilayerconvolutional neural network is then jointly trained based on diagnosticquality images to discriminate the projection quality. The test phase performsfeed-forward propagation over the generator network that demands a very lowcomputational overhead. Extensive evaluations are performed on a largecontrast-enhanced MR dataset of pediatric patients. In particular, images ratedbased on expert radiologists corroborate that GANCS retrieves high contrastimages with detailed texture relative to conventional CS, and pixel-wiseschemes. In addition, it offers reconstruction under a few milliseconds, twoorders of magnitude faster than state-of-the-art CS-MRI schemes.

Deep Generative Adversarial Networks for Compressed Sensing Automates MRI