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

Most current approaches to undersampled multi-coil MRI reconstruction focuson learning the reconstruction model for a fixed, equidistant acquisitiontrajectory. In this paper, we study the problem of joint learning of thereconstruction model together with acquisition policies. To this end, we extendthe End-to-End Variational Network with learnable acquisition policies that canadapt to different data points. We validate our model on a coil-compressedversion of the large scale undersampled multi-coil fastMRI dataset using twoundersampling factors: $4\times$ and $8\times$. Our experiments show on-parperformance with the learnable non-adaptive and handcrafted equidistantstrategies at $4\times$, and an observed improvement of more than $2\%$ in SSIMat $8\times$ acceleration, suggesting that potentially-adaptive $k$-spaceacquisition trajectories can improve reconstructed image quality for largeracceleration factors. However, and perhaps surprisingly, our best performingpolicies learn to be explicitly non-adaptive.

On learning adaptive acquisition policies for undersampled multi-coil MRI reconstruction