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

Patient-specific 3D printing of congenital heart anatomy demands an accuratesegmentation of the thin tissue interfaces which characterise these diagnoses.Even when a label set has a high spatial overlap with the ground truth,inaccurate delineation of these interfaces can result in topological errors.These compromise the clinical utility of such models due to the anomalousappearance of defects. CNNs have achieved state-of-the-art performance insegmentation tasks. Whilst data augmentation has often played an importantrole, we show that conventional image resampling schemes used therein canintroduce topological changes in the ground truth labelling of augmentedsamples. We present a novel pipeline to correct for these changes, using afast-marching algorithm to enforce the topology of the ground truth labelswithin their augmented representations. In so doing, we invoke the idea ofcardiac contiguous topology to describe an arbitrary combination of congenitalheart defects and develop an associated, clinically meaningful metric tomeasure the topological correctness of segmentations. In a series of five-foldcross-validations, we demonstrate the performance gain produced by thispipeline and the relevance of topological considerations to the segmentation ofcongenital heart defects. We speculate as to the applicability of this approachto any segmentation task involving morphologically complex targets.

Topology-preserving augmentation for CNN-based segmentation of congenital heart defects from 3D paediatric CMR