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

Currently, most bone implants used in orthopedics and traumatology arenon-degradable and may need to be surgically removed later on e.g. in the caseof children. This removal is associated with health risks which could beminimized by using biodegradable implants. Therefore, research onmagnesium-based implants is ongoing, which can be objectively quantifiedthrough synchrotron radiation microtomography and subsequent image analysis. Inorder to evaluate the suitability of these materials, e.g. their stability overtime, accurate pixelwise segmentations of these high-resolution scans arenecessary. The fully-convolutional U-Net architecture achieves a Dicecoefficient of 0.750 +/- 0.102 when trained with a small dataset with denseexpert annotations. However, extending the learning to larger databases wouldrequire prohibitive annotation efforts. Hence, in this work we implemented andcompared new training methods that require only a small fraction of manuallyannotated pixels. While directly training on these scribble annotationdeteriorates the segmentation quality by 26.8 percentage points, our new randomwalk-based semi-automatic target achieves the same Dice overlap as a densesupervision, and thus offers a more promising approach for sparse annotations.

Sparse Annotations with Random Walks for U-Net Segmentation of Biodegradable Bone Implants in Synchrotron Microtomograms