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

We present a Three-level Hierarchical Transformer Network (3-level-HTN) formodeling long-term dependencies across clinical notes for the purpose ofpatient-level prediction. The network is equipped with three levels ofTransformer-based encoders to learn progressively from words to sentences,sentences to notes, and finally notes to patients. The first level from word tosentence directly applies a pre-trained BERT model as a fully trainablecomponent. While the second and third levels both implement a stack oftransformer-based encoders, before the final patient representation is fed intoa classification layer for clinical predictions. Compared to conventional BERTmodels, our model increases the maximum input length from 512 tokens to muchlonger sequences that are appropriate for modeling large numbers of clinicalnotes. We empirically examine different hyper-parameters to identify an optimaltrade-off given computational resource limits. Our experiment results on theMIMIC-III dataset for different prediction tasks demonstrate that the proposedHierarchical Transformer Network outperforms previous state-of-the-art models,including but not limited to BigBird.

Three-level Hierarchical Transformer Networks for Long-sequence and Multiple Clinical Documents Classification