Structured LLM Augmentation for Clinical Information Extraction...20th World Congress on Medical and Health Informatics, Aug 09 - 13, 2025, Taipei, Taiwan.
Information extraction tasks, such as Named Entity Recognition (NER) and Relation Extraction (RE), are essential for advancing clinical research and applications. However, these tasks are hindered by the scarcity of labeled clinical documents due to privacy concerns and high annotation costs. This s...
| Publicado en: | Studies in Health Technology & Informatics Vol. 329; pp. 971 - 977 |
|---|---|
| Autores principales: | , , |
| Formato: | proceedings research tables/charts Journal Article |
| Publicado: |
Sage Publications Inc.
2025
|
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=187335002&site=ehost-live header: @attributes: shortDbName: ccm uiTerm: 187335002 longDbName: CINAHL Complete uiTag: AN controlInfo: bkinfo: dissinfo: jinfo: jid: 09269630 U1V jtl: Studies in Health Technology & Informatics issn: 09269630 maglogo: N pubinfo: dt: 2025 vid: 329 pid: 344 pub: Sage Publications Inc. place: Thousand Oaks, California artinfo: ui: 187335002 187335002 187335002 10.3233/SHTI250984 187335002 ppf: 971 ppct: 6 formats: tig: atl: Structured LLM Augmentation for Clinical Information Extraction...20th World Congress on Medical and Health Informatics, Aug 09 - 13, 2025, Taipei, Taiwan. aug: au: Ying Wei Qi Li Pillai, Jay affil: Iowa State University sug: subj: Natural Language Processing Electronic Health Records Clinical Information Systems Task Performance and Analysis Conceptual Framework Models, Theoretical Data Mining Methods Information Retrieval Methods Human Funding Source Congresses and Conferences Taiwan Taiwan Programming Languages Medical Informatics Descriptive Statistics Artificial Intelligence Bioinformatics Decision Support Systems, Clinical Patient Record Systems ab: Information extraction tasks, such as Named Entity Recognition (NER) and Relation Extraction (RE), are essential for advancing clinical research and applications. However, these tasks are hindered by the scarcity of labeled clinical documents due to privacy concerns and high annotation costs. This study introduces a novel framework combining Large Language Models (LLMs) for data augmentation with an adapted BERT model for clinical information extraction. The framework encodes entity and relational information within clinical note segments, enabling LLMs to generate diverse and contextually accurate augmentations while preserving structural integrity. Augmented data is used to train a segmentationbased BERT model, overcoming sequence length limitations and integrating global context via BiLSTM. Evaluations on public and proprietary datasets demonstrate significant performance improvements, highlighting the approach's potential to address data scarcity in clinical information extraction tasks. pubtype: Academic Journal doctype: proceedings research tables/charts Journal Article ougenre: Article language: English refInfo: holdings: @attributes: islocal: N |
|---|