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

Electronic health record (EHR) data are increasingly used to supportreal-world evidence (RWE) studies. Yet its ability to generate reliable RWE islimited by the lack of readily available precise information on the timing ofclinical events such as the onset time of heart failure. We propose aLAbel-efficienT incidenT phEnotyping (LATTE) algorithm to accurately annotatethe timing of clinical events from longitudinal EHR data. By leveraging thepre-trained semantic embedding vectors from large-scale EHR data as priorknowledge, LATTE selects predictive EHR features in a concept re-weightingmodule by mining their relationship to the target event and compresses theirinformation into longitudinal visit embeddings through a visit attentionlearning network. LATTE employs a recurrent neural network to capture thesequential dependency between the target event and visit embeddingsbefore/after it. To improve label efficiency, LATTE constructs highlyinformative longitudinal silver-standard labels from large-scale unlabeledpatients to perform unsupervised pre-training and semi-supervised jointtraining. Finally, LATTE enhances cross-site portability via contrastiverepresentation learning. LATTE is evaluated on three analyses: the onset oftype-2 diabetes, heart failure, and the onset and relapses of multiplesclerosis. We use various evaluation metrics present in the literatureincluding the $ABC_{gain}$, the proportion of reduction in the area between theobserved event indicator and the predicted cumulative incidences in referenceto the prediction per incident prevalence. LATTE consistently achievessubstantial improvement over benchmark methods such as SAMGEP and RETAIN in allsettings.

LATTE: Label-efficient Incident Phenotyping from Longitudinal Electronic Health Records