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

Recently, deep learning approaches for MR motion artifact correction havebeen extensively studied. Although these approaches have shown high performanceand reduced computational complexity compared to classical methods, most ofthem require supervised training using paired artifact-free andartifact-corrupted images, which may prohibit its use in many importantclinical applications. For example, transient severe motion (TSM) due to acutetransient dyspnea in Gd-EOB-DTPA-enhanced MR is difficult to control and modelfor paired data generation. To address this issue, here we propose a novelunsupervised deep learning scheme through outlier-rejecting bootstrapsubsampling and aggregation. This is inspired by the observation that motionsusually cause sparse k-space outliers in the phase encoding direction, sok-space subsampling along the phase encoding direction can remove some outliersand the aggregation step can further improve the results from thereconstruction network. Our method does not require any paired data because thetraining step only requires artifact-free images. Furthermore, to address thesmoothing from potential bias to the artifact-free images, the network istrained in an unsupervised manner using optimal transport driven cycleGAN. Weverify that our method can be applied for artifact correction from simulatedmotion as well as real motion from TSM successfully, outperforming existingstate-of-the-art deep learning methods.

Unsupervised MR Motion Artifact Deep Learning using Outlier-Rejecting Bootstrap Aggregation