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

Undersampling k-space data in MRI reduces scan time but pose challenges inimage reconstruction. Considerable progress has been made in reconstructingaccelerated MRI. However, restoration of high-frequency image details in highlyundersampled data remains challenging. To address this issue, we proposeCAMP-Net, an unrolling-based Consistency-Aware Multi-Prior Network foraccelerated MRI reconstruction. CAMP-Net leverages complementary multi-priorknowledge and multi-slice information from various domains to enhancereconstruction quality. Specifically, CAMP-Net comprises three interleavedmodules for image enhancement, k-space restoration, and calibrationconsistency, respectively. These modules jointly learn priors from data inimage domain, k-domain, and calibration region, respectively, in data-drivenmanner during each unrolled iteration. Notably, the encoded calibration priorknowledge extracted from auto-calibrating signals implicitly guides thelearning of consistency-aware k-space correlation for reliable interpolation ofmissing k-space data. To maximize the benefits of image domain and k-domainprior knowledge, the reconstructions are aggregated in a frequency fusionmodule, exploiting their complementary properties to optimize the trade-offbetween artifact removal and fine detail preservation. Additionally, weincorporate a surface data fidelity layer during the learning of k-domain andcalibration domain priors to prevent degradation of the reconstruction causedby padding-induced data imperfections. We evaluate the generalizability androbustness of our method on three large public datasets with varyingacceleration factors and sampling patterns. The experimental resultsdemonstrate that our method outperforms state-of-the-art approaches in terms ofboth reconstruction quality and $T_2$ mapping estimation, particularly inscenarios with high acceleration factors.

CAMP-Net: Consistency-Aware Multi-Prior Network for Accelerated MRI Reconstruction