A Hybrid Natural Language Processing Platform for Multi-Site RWD Studies...20th World Congress on Medical and Health Informatics (MEDINFO), August 9-13, 2025, Taipei, Taiwan.

Real-world data (RWD) obtained from electronic medical records has become a valuable resource for healthcare research. However, integrating unstructured free-text clinical data remains a significant challenge. Although natural language processing (NLP) offers a promising solution, its implementation...

Descripción completa

Detalles Bibliográficos
Publicado en:Studies in Health Technology & Informatics Vol. 329; pp. 1230 - 1235
Autores principales: Kento SUGIMOTO, Yasushi MATSUMURA, Shoya WADA, Shozo KONISHI, Katsuki OKADA, Toshihiro TAKEDA
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=187335053&site=ehost-live
header:
  @attributes:
    shortDbName: ccm
    uiTerm: 187335053
    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:
        187335053
        187335053
        187335053
        10.3233/SHTI251035
        187335053
      ppf: 1230
      ppct: 5
      formats:
      tig:
        atl: A Hybrid Natural Language Processing Platform for Multi-Site RWD Studies...20th World Congress on Medical and Health Informatics (MEDINFO), August 9-13, 2025, Taipei, Taiwan.
      aug:
        au:
          Kento SUGIMOTO
          Yasushi MATSUMURA
          Shoya WADA
          Shozo KONISHI
          Katsuki OKADA
          Toshihiro TAKEDA
        affil: Department of Medical Informatics, Osaka University Graduate School of Medicine, Osaka, Japan
      sug:
        subj:
          Natural Language Processing
          Electronic Health Records
          Data Security
          Information Retrieval
          Human
          Congresses and Conferences Taiwan
          Taiwan
          Deep Learning
          Workflow
          Funding Source
      ab: Real-world data (RWD) obtained from electronic medical records has become a valuable resource for healthcare research. However, integrating unstructured free-text clinical data remains a significant challenge. Although natural language processing (NLP) offers a promising solution, its implementation is frequently hampered by high computational costs. Moreover, privacy concerns complicate data integration in multi-site RWD studies. This study proposes a hybrid platform that integrates centralized NLP processing with robust privacy protection, facilitating effective information extraction from free-text data across various institutions. We performed comparative experiments utilizing 500 sample reports to assess the efficacy of the proposed hybrid platform against a fully distributed method using on-site servers. The results indicated that the central graphics processing units server significantly outperformed the site central processing units, processing reports in 0.12 s compared to an average of 64.23 s. Additionally, the central server exhibited a low and consistent increase in processing time regardless of report lengths, highlighting its efficiency and scalability. Our developed hybrid platform enhances computational efficiency while tackling privacy and data governance issues.
      pubtype: Academic Journal
      doctype:
        proceedings
        research
        tables/charts
        Journal Article
      ougenre: Article
    language: English
    refInfo:
    holdings:
      @attributes:
        islocal: N