Literature DB >> 28267590

Unsupervised ensemble ranking of terms in electronic health record notes based on their importance to patients.

Jinying Chen1, Hong Yu2.   

Abstract

BACKGROUND: Allowing patients to access their own electronic health record (EHR) notes through online patient portals has the potential to improve patient-centered care. However, EHR notes contain abundant medical jargon that can be difficult for patients to comprehend. One way to help patients is to reduce information overload and help them focus on medical terms that matter most to them. Targeted education can then be developed to improve patient EHR comprehension and the quality of care.
OBJECTIVE: The aim of this work was to develop FIT (Finding Important Terms for patients), an unsupervised natural language processing (NLP) system that ranks medical terms in EHR notes based on their importance to patients.
METHODS: We built FIT on a new unsupervised ensemble ranking model derived from the biased random walk algorithm to combine heterogeneous information resources for ranking candidate terms from each EHR note. Specifically, FIT integrates four single views (rankers) for term importance: patient use of medical concepts, document-level term salience, word co-occurrence based term relatedness, and topic coherence. It also incorporates partial information of term importance as conveyed by terms' unfamiliarity levels and semantic types. We evaluated FIT on 90 expert-annotated EHR notes and used the four single-view rankers as baselines. In addition, we implemented three benchmark unsupervised ensemble ranking methods as strong baselines.
RESULTS: FIT achieved 0.885 AUC-ROC for ranking candidate terms from EHR notes to identify important terms. When including term identification, the performance of FIT for identifying important terms from EHR notes was 0.813 AUC-ROC. Both performance scores significantly exceeded the corresponding scores from the four single rankers (P<0.001). FIT also outperformed the three ensemble rankers for most metrics. Its performance is relatively insensitive to its parameter.
CONCLUSIONS: FIT can automatically identify EHR terms important to patients. It may help develop future interventions to improve quality of care. By using unsupervised learning as well as a robust and flexible framework for information fusion, FIT can be readily applied to other domains and applications.
Copyright © 2017 Elsevier Inc. All rights reserved.

Entities:  

Keywords:  Electronic health record; Information extraction; Natural language processing; Unsupervised ensemble ranking

Mesh:

Year:  2017        PMID: 28267590      PMCID: PMC5505865          DOI: 10.1016/j.jbi.2017.02.016

Source DB:  PubMed          Journal:  J Biomed Inform        ISSN: 1532-0464            Impact factor:   6.317


  41 in total

1.  Characteristics of consumer terminology for health information retrieval.

Authors:  Q Zeng; S Kogan; N Ash; R A Greenes; A A Boxwala
Journal:  Methods Inf Med       Date:  2002       Impact factor: 2.176

2.  Identifying consumer-friendly display (CFD) names for health concepts.

Authors:  Qing T Zeng; Tony Tse; Jon Crowell; Guy Divita; Laura Roth; Allen C Browne
Journal:  AMIA Annu Symp Proc       Date:  2005

3.  Identifying important concepts from medical documents.

Authors:  Quanzhi Li; Yi-Fang Brook Wu
Journal:  J Biomed Inform       Date:  2006-03-02       Impact factor: 6.317

4.  The Medicare Electronic Health Record Incentive Program: provider performance on core and menu measures.

Authors:  Adam Wright; Joshua Feblowitz; Lipika Samal; Allison B McCoy; Dean F Sittig
Journal:  Health Serv Res       Date:  2013-12-21       Impact factor: 3.402

5.  Opportunities to improve clinical summaries for patients at hospital discharge.

Authors:  Erin Sarzynski; Hamza Hashmi; Jeevarathna Subramanian; Laurie Fitzpatrick; Molly Polverento; Michael Simmons; Kevin Brooks; Charles Given
Journal:  BMJ Qual Saf       Date:  2016-05-06       Impact factor: 7.035

6.  Readability of patient education materials on the American Orthopaedic Society for Sports Medicine website.

Authors:  Adam E M Eltorai; Alex Han; Jeremy Truntzer; Alan H Daniels
Journal:  Phys Sportsmed       Date:  2014-11       Impact factor: 2.241

7.  Readability assessment of internet-based consumer health information.

Authors:  Tiffany M Walsh; Teresa A Volsko
Journal:  Respir Care       Date:  2008-10       Impact factor: 2.258

8.  Lay understanding of terms used in cancer consultations.

Authors:  Kristina Chapman; Charles Abraham; Valerie Jenkins; Lesley Fallowfield
Journal:  Psychooncology       Date:  2003-09       Impact factor: 3.894

9.  Bidirectional RNN for Medical Event Detection in Electronic Health Records.

Authors:  Abhyuday N Jagannatha; Hong Yu
Journal:  Proc Conf       Date:  2016-06

10.  Patient experiences with full electronic access to health records and clinical notes through the My HealtheVet Personal Health Record Pilot: qualitative study.

Authors:  Susan S Woods; Erin Schwartz; Anais Tuepker; Nancy A Press; Kim M Nazi; Carolyn L Turvey; W Paul Nichol
Journal:  J Med Internet Res       Date:  2013-03-27       Impact factor: 5.428

View more
  3 in total

1.  Interventions to increase patient portal use in vulnerable populations: a systematic review.

Authors:  Lisa V Grossman; Ruth M Masterson Creber; Natalie C Benda; Drew Wright; David K Vawdrey; Jessica S Ancker
Journal:  J Am Med Inform Assoc       Date:  2019-08-01       Impact factor: 4.497

Review 2.  Natural Language Processing for EHR-Based Computational Phenotyping.

Authors:  Zexian Zeng; Yu Deng; Xiaoyu Li; Tristan Naumann; Yuan Luo
Journal:  IEEE/ACM Trans Comput Biol Bioinform       Date:  2018-06-25       Impact factor: 3.710

3.  Ranking Medical Terms to Support Expansion of Lay Language Resources for Patient Comprehension of Electronic Health Record Notes: Adapted Distant Supervision Approach.

Authors:  Jinying Chen; Abhyuday N Jagannatha; Samah J Fodeh; Hong Yu
Journal:  JMIR Med Inform       Date:  2017-10-31
  3 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.