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

Fetal pose estimation in 3D ultrasound (US) involves identifying a set ofassociated fetal anatomical landmarks. Its primary objective is to providecomprehensive information about the fetus through landmark connections, thusbenefiting various critical applications, such as biometric measurements, planelocalization, and fetal movement monitoring. However, accurately estimating the3D fetal pose in US volume has several challenges, including poor imagequality, limited GPU memory for tackling high dimensional data, symmetrical orambiguous anatomical structures, and considerable variations in fetal poses. Inthis study, we propose a novel 3D fetal pose estimation framework (calledFetusMapV2) to overcome the above challenges. Our contribution is three-fold.First, we propose a heuristic scheme that explores the complementary networkstructure-unconstrained and activation-unreserved GPU memory managementapproaches, which can enlarge the input image resolution for better resultsunder limited GPU memory. Second, we design a novel Pair Loss to mitigateconfusion caused by symmetrical and similar anatomical structures. It separatesthe hidden classification task from the landmark localization task and thusprogressively eases model learning. Last, we propose a shape priors-basedself-supervised learning by selecting the relatively stable landmarks to refinethe pose online. Extensive experiments and diverse applications on alarge-scale fetal US dataset including 1000 volumes with 22 landmarks pervolume demonstrate that our method outperforms other strong competitors.

FetusMapV2: Enhanced Fetal Pose Estimation in 3D Ultrasound