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

Various deep learning models have been proposed for 3D bone shapereconstruction from two orthogonal (biplanar) X-ray images. However, it isunclear how these models compare against each other since they are evaluated ondifferent anatomy, cohort and (often privately held) datasets. Moreover, theimpact of the commonly optimized image-based segmentation metrics such as dicescore on the estimation of clinical parameters relevant in 2D-3D bone shapereconstruction is not well known. To move closer toward clinical translation,we propose a benchmarking framework that evaluates tasks relevant to real-worldclinical scenarios, including reconstruction of fractured bones, bones withimplants, robustness to population shift, and error in estimating clinicalparameters. Our open-source platform provides reference implementations of 8models (many of whose implementations were not publicly available), APIs toeasily collect and preprocess 6 public datasets, and the implementation ofautomatic clinical parameter and landmark extraction methods. We present anextensive evaluation of 8 2D-3D models on equal footing using 6 public datasetscomprising images for four different anatomies. Our results show thatattention-based methods that capture global spatial relationships tend toperform better across all anatomies and datasets; performance on clinicallyrelevant subgroups may be overestimated without disaggregated reporting; ribsare substantially more difficult to reconstruct compared to femur, hip andspine; and the dice score improvement does not always bring a correspondingimprovement in the automatic estimation of clinically relevant parameters.

Benchmarking Encoder-Decoder Architectures for Biplanar X-ray to 3D Shape Reconstruction