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

Current approaches for clinical information extraction are inefficient interms of computational costs and memory consumption, hindering theirapplication to process large-scale electronic health records (EHRs). We proposean efficient end-to-end model, the Joint-NER-RE-Fourier (JNRF), to jointlylearn the tasks of named entity recognition and relation extraction fordocuments of variable length. The architecture uses positional encoding andunitary batch sizes to process variable length documents and uses aweight-shared Fourier network layer for low-complexity token mixing. Finally,we reach the theoretical computational complexity lower bound for relationextraction using a selective pooling strategy and distance-aware attentionweights with trainable polynomial distance functions. We evaluated the JNRFarchitecture using the 2018 N2C2 ADE benchmark to jointly extractmedication-related entities and relations in variable-length EHR summaries.JNRF outperforms rolling window BERT with selective pooling by 0.42%, whilebeing twice as fast to train. Compared to state-of-the-art BiLSTM-CRFarchitectures on the N2C2 ADE benchmark, results show that the proposedapproach trains 22 times faster and reduces GPU memory consumption by 1.75folds, with a reasonable performance tradeoff of 90%, without the use ofexternal tools, hand-crafted rules or post-processing. Given the significantcarbon footprint of deep learning models and the current energy crises, thesemethods could support efficient and cleaner information extraction in EHRs andother types of large-scale document databases.

Efficient Joint Learning for Clinical Named Entity Recognition and Relation Extraction Using Fourier Networks: A Use Case in Adverse Drug Events