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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Fluoroscopy is the standard imaging modality used to guide hip surgery and istherefore a natural sensor for computer-assisted navigation. In order toefficiently solve the complex registration problems presented duringnavigation, human-assisted annotations of the intraoperative image aretypically required. This manual initialization interferes with the surgicalworkflow and diminishes any advantages gained from navigation. We propose amethod for fully automatic registration using annotations produced by a neuralnetwork. Neural networks are trained to simultaneously segment anatomy andidentify landmarks in fluoroscopy. Training data is obtained using anintraoperatively incompatible 2D/3D registration of hip anatomy. Ground truth2D labels are established using projected 3D annotations. Intraoperativeregistration couples an intensity-based strategy with annotations inferred bythe network and requires no human assistance. Ground truth labels were obtainedin 366 fluoroscopic images across 6 cadaveric specimens. In aleave-one-subject-out experiment, networks obtained mean dice coefficients forleft and right hemipelves, left and right femurs of 0.86, 0.87, 0.90, and 0.84.The mean 2D landmark error was 5.0 mm. The pelvis was registered within 1degree for 86% of the images when using the proposed intraoperative approachwith an average runtime of 7 seconds. In comparison, an intensity-only approachwithout manual initialization, registered the pelvis to 1 degree in 18% ofimages. We have created the first accurately annotated, non-synthetic, datasetof hip fluoroscopy. By using these annotations as training data for neuralnetworks, state of the art performance in fluoroscopic segmentation andlandmark localization was achieved. Integrating these annotations allows for arobust, fully automatic, and efficient intraoperative registration duringfluoroscopic navigation of the hip.

Automatic Annotation of Hip Anatomy in Fluoroscopy for Robust and Efficient 2D/3D Registration