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

Synthesizing medical images while preserving their structural information iscrucial in medical research. In such scenarios, the preservation of anatomicalcontent becomes especially important. Although recent advances have been madeby incorporating instance-level information to guide translation, these methodsoverlook the spatial coherence of structural-level representation and theanatomical invariance of content during translation. To address these issues,we introduce hierarchical granularity discrimination, which exploits variouslevels of semantic information present in medical images. Our strategy utilizesthree levels of discrimination granularity: pixel-level discrimination using aBrain Memory Bank, structure-level discrimination on each brain structure witha re-weighting strategy to focus on hard samples, and global-leveldiscrimination to ensure anatomical consistency during translation. The imagetranslation performance of our strategy has been evaluated on three independentdatasets (UK Biobank, IXI, and BraTS 2018), and it has outperformedstate-of-the-art algorithms. Particularly, our model excels not only insynthesizing normal structures but also in handling abnormal (pathological)structures, such as brain tumors, despite the variations in contrast observedacross different imaging modalities due to their pathological characteristics.The diagnostic value of synthesized MR images containing brain tumors has beenevaluated by radiologists. This indicates that our model may offer analternative solution in scenarios where specific MR modalities of patients areunavailable. Extensive experiments further demonstrate the versatility of ourmethod, providing unique insights into medical image translation.

HiFi-Syn: Hierarchical Granularity Discrimination for High-Fidelity Synthesis of MR Images with Structure Preservation