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

Efficiently quantifying renal structures can provide distinct spatial contextand facilitate biomarker discovery for kidney morphology. However, thedevelopment and evaluation of the transformer model to segment the renalcortex, medulla, and collecting system remains challenging due to datainefficiency. Inspired by the hierarchical structures in vision transformer, wepropose a novel method using a 3D block aggregation transformer for segmentingkidney components on contrast-enhanced CT scans. We construct the first cohortof renal substructures segmentation dataset with 116 subjects underinstitutional review board (IRB) approval. Our method yields thestate-of-the-art performance (Dice of 0.8467) against the baseline approach of0.8308 with the data-efficient design. The Pearson R achieves 0.9891 betweenthe proposed method and manual standards and indicates the strong correlationand reproducibility for volumetric analysis. We extend the proposed method tothe public KiTS dataset, the method leads to improved accuracy compared totransformer-based approaches. We show that the 3D block aggregation transformercan achieve local communication between sequence representations withoutmodifying self-attention, and it can serve as an accurate and efficientquantification tool for characterizing renal structures.

Characterizing Renal Structures with 3D Block Aggregate Transformers