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

Deep learning (DL) based predictive models from electronic health records(EHR) deliver impressive performance in many clinical tasks. Large trainingcohorts, however, are often required to achieve high accuracy, hindering theadoption of DL-based models in scenarios with limited training data size.Recently, bidirectional encoder representations from transformers (BERT) andrelated models have achieved tremendous successes in the natural languageprocessing domain. The pre-training of BERT on a very large training corpusgenerates contextualized embeddings that can boost the performance of modelstrained on smaller datasets. We propose Med-BERT, which adapts the BERTframework for pre-training contextualized embedding models on structureddiagnosis data from 28,490,650 patients EHR dataset. Fine-tuning experimentsare conducted on two disease-prediction tasks: (1) prediction of heart failurein patients with diabetes and (2) prediction of pancreatic cancer from twoclinical databases. Med-BERT substantially improves prediction accuracy,boosting the area under receiver operating characteristics curve (AUC) by2.02-7.12%. In particular, pre-trained Med-BERT substantially improves theperformance of tasks with very small fine-tuning training sets (300-500samples) boosting the AUC by more than 20% or equivalent to the AUC of 10 timeslarger training set. We believe that Med-BERT will benefit disease-predictionstudies with small local training datasets, reduce data collection expenses,and accelerate the pace of artificial intelligence aided healthcare.

Med-BERT: pre-trained contextualized embeddings on large-scale structured electronic health records for disease prediction