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/

Zendy Plus

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

Zendy Tools

Zendy Open

In cardiac CINE, motion-compensated MR reconstruction (MCMR) is an effectiveapproach to address highly undersampled acquisitions by incorporating motioninformation between frames. In this work, we propose a novel perspective foraddressing the MCMR problem and a more integrated and efficient solution to theMCMR field. Contrary to state-of-the-art (SOTA) MCMR methods which break theoriginal problem into two sub-optimization problems, i.e. motion estimation andreconstruction, we formulate this problem as a single entity with one singleoptimization. Our approach is unique in that the motion estimation is directlydriven by the ultimate goal, reconstruction, but not by the canonicalmotion-warping loss (similarity measurement between motion-warped images andtarget images). We align the objectives of motion estimation andreconstruction, eliminating the drawbacks of artifacts-affected motionestimation and therefore error-propagated reconstruction. Further, we candeliver high-quality reconstruction and realistic motion without applying anyregularization/smoothness loss terms, circumventing the non-trivial weightingfactor tuning. We evaluate our method on two datasets: 1) an in-house acquired2D CINE dataset for the retrospective study and 2) the public OCMR cardiacdataset for the prospective study. The conducted experiments indicate that theproposed MCMR framework can deliver artifact-free motion estimation andhigh-quality MR images even for imaging accelerations up to 20x, outperformingSOTA non-MCMR and MCMR methods in both qualitative and quantitative evaluationacross all experiments. The code is available athttps://github.com/JZPeterPan/MCMR-Recon-Driven-Motion.

Motion-compensated MR CINE reconstruction with reconstruction-driven motion estimation