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

Accelerated MRI protocols routinely involve a predefined sampling patternthat undersamples the k-space. Finding an optimal pattern can enhance thereconstruction quality, however this optimization is a challenging task. Toaddress this challenge, we introduce a novel deep learning framework, AutoSamp,based on variational information maximization that enables joint optimizationof sampling pattern and reconstruction of MRI scans. We represent the encoderas a non-uniform Fast Fourier Transform that allows continuous optimization ofk-space sample locations on a non-Cartesian plane, and the decoder as a deepreconstruction network. Experiments on public MRI datasets show improvedreconstruction quality of the proposed AutoSamp method over the prevailingvariable density and variable density Poisson disc sampling. We demonstratethat our data-driven sampling optimization method achieves 4.4dB, 2.0dB,0.75dB, 0.7dB PSNR improvements over reconstruction with Poisson Disc masks foracceleration factors of R = 5, 10, 15, 25, respectively. Furthermore, weanalyze the characteristics of the learned sampling patterns with respect tochanges in acceleration factor, measurement noise, underlying anatomy, and coilsensitivities. We show that all these factors contribute to the optimizationresult by affecting the sampling density, k-space coverage and point spreadfunctions of the learned sampling patterns.

AutoSamp: Autoencoding MRI Sampling via Variational Information Maximization