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

Sifting through vast textual data and summarizing key information fromelectronic health records (EHR) imposes a substantial burden on how cliniciansallocate their time. Although large language models (LLMs) have shown immensepromise in natural language processing (NLP) tasks, their efficacy on a diverserange of clinical summarization tasks has not yet been rigorously demonstrated.In this work, we apply domain adaptation methods to eight LLMs, spanning sixdatasets and four distinct clinical summarization tasks: radiology reports,patient questions, progress notes, and doctor-patient dialogue. Our thoroughquantitative assessment reveals trade-offs between models and adaptationmethods in addition to instances where recent advances in LLMs may not improveresults. Further, in a clinical reader study with ten physicians, we show thatsummaries from our best-adapted LLMs are preferable to human summaries in termsof completeness and correctness. Our ensuing qualitative analysis highlightschallenges faced by both LLMs and human experts. Lastly, we correlatetraditional quantitative NLP metrics with reader study scores to enhance ourunderstanding of how these metrics align with physician preferences. Ourresearch marks the first evidence of LLMs outperforming human experts inclinical text summarization across multiple tasks. This implies thatintegrating LLMs into clinical workflows could alleviate documentation burden,empowering clinicians to focus more on personalized patient care and theinherently human aspects of medicine.

Clinical Text Summarization: Adapting Large Language Models Can Outperform Human Experts