Identification of Dietary Supplement Use from Electronic Health Records Using Transformer-based Language Models

被引:2
作者
Zhou, Sicheng [1 ]
Schutte, Dalton
Xing, Aiwen [2 ]
Chen, Jiyang [3 ]
Wolfson, Julian [4 ]
He, Zhe [5 ]
Yu, Fang [6 ]
Zhang, Rui [1 ,7 ]
机构
[1] Univ Minnesota, Inst Hlth Informat, Minneapolis, MN 55446 USA
[2] Florida State Univ, Dept Stat, Tallahassee, FL 32306 USA
[3] Univ Minnesota, Coll Sci & Engn, Minneapolis, MN 55446 USA
[4] Univ Minnesota, Div Biostat, Minneapolis, MN 55446 USA
[5] Florida State Univ, Coll Commun & Informat, Tallahassee, FL 32306 USA
[6] Arizona State Univ, Edson Coll Nursing & Hlth Innovat, Phoenix, AZ 85004 USA
[7] Univ Minnesota, Coll Pharm, Minneapolis, MN 55446 USA
来源
2021 IEEE 9TH INTERNATIONAL CONFERENCE ON HEALTHCARE INFORMATICS (ICHI 2021) | 2021年
关键词
BERT; Natural language processing; Dietary supplements; EHR; MCI; ADRD;
D O I
10.1109/ICHI52183.2021.00096
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The roles of dietary supplement (DS) usage on disease progression of patients with cognitive impairments remain unclear. Transformed-based language models were trained to identify DS use status from clinical notes among patients with Alzheimer's disease and related dementias (ADRD). The best name entity recognition for DS achieved F1-score is 0.964 and the PubMed BERT based use status classifier achieved the weighted F1-score of 0.879. Integrating with DS use from medication table, we identified totally 125 unique DS among patients with mild cognitive impairment (MCI) only and 108 unique DS among patients who progressed to ADRD.
引用
收藏
页码:513 / 514
页数:2
相关论文
共 11 条
  • [1] Alsentzer E, 2019, Arxiv, DOI [arXiv:1904.03323, 10.48550/arXiv.1904.03323, DOI 10.48550/ARXIV.1904.03323]
  • [2] Using word embeddings to expand terminology of dietary supplements on clinical notes
    Fan, Yadan
    Pakhomov, Serguei
    McEwan, Reed
    Zhao, Wendi
    Lindemann, Elizabeth
    Zhang, Rui
    [J]. JAMIA OPEN, 2019, 2 (02) : 246 - 253
  • [3] Using natural language processing methods to classify use status of dietary supplements in clinical notes
    Fan, Yadan
    Zhang, Rui
    [J]. BMC MEDICAL INFORMATICS AND DECISION MAKING, 2018, 18
  • [4] Gu Y., arXiv
  • [5] BioBERT: a pre-trained biomedical language representation model for biomedical text mining
    Lee, Jinhyuk
    Yoon, Wonjin
    Kim, Sungdong
    Kim, Donghyeon
    Kim, Sunkyu
    So, Chan Ho
    Kang, Jaewoo
    [J]. BIOINFORMATICS, 2020, 36 (04) : 1234 - 1240
  • [6] iDISK: the integrated DIetary Supplements Knowledge base
    Rizvi, Rubina F.
    Vasilakes, Jake
    Adam, Terrence J.
    Melton, Genevieve B.
    Bishop, Jeffrey R.
    Bian, Jiang
    Tao, Cui
    Zhang, Rui
    [J]. JOURNAL OF THE AMERICAN MEDICAL INFORMATICS ASSOCIATION, 2020, 27 (04) : 539 - 548
  • [7] Detecting drug-drug interactions using a database for spontaneous adverse drug reactions: an example with diuretics and non-steroidal anti-inflammatory drugs
    van Puijenbroek, EP
    Egberts, ACG
    Heerdink, ER
    Leufkens, HGM
    [J]. EUROPEAN JOURNAL OF CLINICAL PHARMACOLOGY, 2000, 56 (9-10) : 733 - 738
  • [8] Vasilakes J., 2020, Journal of American Medical Informatics Association, DOI DOI 10.1093/JAMIA/OCAA128
  • [9] who.int, WHO DEM
  • [10] World Health O, RISK RED COGN DECL D