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

Though analysis of Medical Images by Deep Learning achieves unprecedentedresults across various applications, the effect of \emph{noisy trainingannotations} is rarely studied in a systematic manner. In Medical ImageAnalysis, most reports addressing this question concentrate on studyingsegmentation performance of deep learning classifiers. The absence ofcontinuous ground truth annotations in these studies limits the value ofconclusions for applications, where regression is the primary method of choice.In the application of surgical instrument pose estimation, where precision hasa direct clinical impact on patient outcome, studying the effect of \emph{noisyannotations} on deep learning pose estimation techniques is of supremeimportance. Real x-ray images are inadequate for this evaluation due to theunavailability of ground truth annotations. We circumvent this problem bygenerating synthetic radiographs, where the ground truth pose is known andtherefore the pose estimation error made by the medical-expert can be estimatedfrom experiments. Furthermore, this study shows the property of deep neuralnetworks to learn dominant signals from noisy annotations with sufficient datain a regression setting.

How Bad is Good enough: Noisy annotations for instrument pose estimation