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

A large and diverse set of measurements are regularly collected during apatient's hospital stay to monitor their health status. Tools for integratingthese measurements into severity scores, that accurately track changes inillness severity, can improve clinicians ability to provide timelyinterventions. Existing approaches for creating such scores either 1) rely onexperts to fully specify the severity score, or 2) train a predictive score,using supervised learning, by regressing against a surrogate marker of severitysuch as the presence of downstream adverse events. The first approach does notextend to diseases where an accurate score cannot be elicited from experts. Thesecond approach often produces scores that suffer from bias due totreatment-related censoring (Paxton, 2013). We propose a novel ranking basedframework for disease severity score learning (DSSL). DSSL exploits thefollowing key observation: while it is challenging for experts to quantify thedisease severity at any given time, it is often easy to compare the diseaseseverity at two different times. Extending existing ranking algorithms, DSSLlearns a function that maps a vector of patient's measurements to a scalarseverity score such that the resulting score is temporally smooth andconsistent with the expert's ranking of pairs of disease states. We apply DSSLto the problem of learning a sepsis severity score using a large, real-worlddataset. The learned scores significantly outperform state-of-the-art clinicalscores in ranking patient states by severity and in early detection of futureadverse events. We also show that the learned disease severity trajectories areconsistent with clinical expectations of disease evolution. Further, usingsimulated datasets, we show that DSSL exhibits better generalizationperformance to changes in treatment patterns compared to the above approaches.

Learning (Predictive) Risk Scores in the Presence of Censoring due to Interventions