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/

Zendy Plus

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

Zendy Tools

Zendy Open

In many screening applications, the primary goal of a radiologist orassisting artificial intelligence is to rule out certain findings. Theclassifiers built for such applications are often trained on large datasetsthat derive labels from clinical notes written for patients. While the qualityof the positive findings described in these notes is often reliable, lack ofthe mention of a finding does not always rule out the presence of it. Thishappens because radiologists comment on the patient in the context of the exam,for example focusing on trauma as opposed to chronic disease at emergencyrooms. However, this disease finding ambiguity can affect the performance ofalgorithms. Hence it is critical to model the ambiguity during training. Wepropose a scheme to apply reasonable class weight modifiers to our lossfunction for the no mention cases during training. We experiment with twodifferent deep neural network architectures and show that the proposed methodresults in a large improvement in the performance of the classifiers, speciallyon negated findings. The baseline performance of a custom-made dilated blocknetwork proposed in this paper shows an improvement in comparison with baselineDenseNet-201, while both architectures benefit from the new proposed lossfunction weighting scheme. Over 200,000 chest X-ray images and three highlycommon diseases, along with their negated counterparts, are included in thisstudy.

Boosting the rule-out accuracy of deep disease detection using class weight modifiers