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

It is often infeasible or impossible to obtain ground truth labels formedical data. To circumvent this, one may build rule-based or otherexpert-knowledge driven labelers to ingest data and yield silver labels absentany ground-truth training data. One popular such labeler is CheXpert, a labelerthat produces diagnostic labels for chest X-ray radiology reports. CheXpert isvery useful, but is relatively computationally slow, especially when integratedwith end-to-end neural pipelines, is non-differentiable so can't be used in anyapplications that require gradients to flow through the labeler, and does notyield probabilistic outputs, which limits our ability to improve the quality ofthe silver labeler through techniques such as active learning.  In this work, we solve all three of these problems with$\texttt{CheXpert++}$, a BERT-based, high-fidelity approximation to CheXpert.$\texttt{CheXpert++}$ achieves 99.81\% parity with CheXpert, which means it canbe reliably used as a drop-in replacement for CheXpert, all while beingsignificantly faster, fully differentiable, and probabilistic in output. Erroranalysis of $\texttt{CheXpert++}$ also demonstrates that $\texttt{CheXpert++}$has a tendency to actually correct errors in the CheXpert labels, with$\texttt{CheXpert++}$ labels being more often preferred by a clinician overCheXpert labels (when they disagree) on all but one disease task. To furtherdemonstrate the utility of these advantages in this model, we conduct aproof-of-concept active learning study, demonstrating we can improve accuracyon an expert labeled random subset of report sentences by approximately 8\%over raw, unaltered CheXpert by using one-iteration of active-learning inspiredre-training. These findings suggest that simple techniques in co-learning andactive learning can yield high-quality labelers under minimal, and controllablehuman labeling demands.

CheXpert++: Approximating the CheXpert labeler for Speed,Differentiability, and Probabilistic Output