JMIR Med Inform. 2016 Nov 30;4(4):e40. doi: 10.2196/medinform.6373.
Finding Important Terms for Patients in Their Electronic Health Records: A Learning-to-Rank Approach Using Expert Annotations.
JMIR medical informatics
Jinying Chen, Jiaping Zheng, Hong Yu
Affiliations
Affiliations
- Department of Quantitative Health Sciences, University of Massachusetts Medical School, Worcester, MA, United States.
- School of Computer Science, University of Massachusetts, Amherst, MA, United States.
- Bedford Veterans Affairs Medical Center, Center for Healthcare Organization and Implementation Research, Bedford, MA, United States.
PMID: 27903489
PMCID: PMC5156821 DOI: 10.2196/medinform.6373
Abstract
BACKGROUND: Many health organizations allow patients to access their own electronic health record (EHR) notes through online patient portals as a way to enhance patient-centered care. However, EHR notes are typically long and contain abundant medical jargon that can be difficult for patients to understand. In addition, many medical terms in patients' notes are not directly related to their health care needs. One way to help patients better comprehend their own notes is to reduce information overload and help them focus on medical terms that matter most to them. Interventions can then be developed by giving them targeted education to improve their EHR comprehension and the quality of care.
OBJECTIVE: We aimed to develop a supervised natural language processing (NLP) system called Finding impOrtant medical Concepts most Useful to patientS (FOCUS) that automatically identifies and ranks medical terms in EHR notes based on their importance to the patients.
METHODS: First, we built an expert-annotated corpus. For each EHR note, 2 physicians independently identified medical terms important to the patient. Using the physicians' agreement as the gold standard, we developed and evaluated FOCUS. FOCUS first identifies candidate terms from each EHR note using MetaMap and then ranks the terms using a support vector machine-based learn-to-rank algorithm. We explored rich learning features, including distributed word representation, Unified Medical Language System semantic type, topic features, and features derived from consumer health vocabulary. We compared FOCUS with 2 strong baseline NLP systems.
RESULTS: Physicians annotated 90 EHR notes and identified a mean of 9 (SD 5) important terms per note. The Cohen's kappa annotation agreement was .51. The 10-fold cross-validation results show that FOCUS achieved an area under the receiver operating characteristic curve (AUC-ROC) of 0.940 for ranking candidate terms from EHR notes to identify important terms. When including term identification, the performance of FOCUS for identifying important terms from EHR notes was 0.866 AUC-ROC. Both performance scores significantly exceeded the corresponding baseline system scores (P<.001). Rich learning features contributed to FOCUS's performance substantially.
CONCLUSIONS: FOCUS can automatically rank terms from EHR notes based on their importance to patients. It may help develop future interventions that improve quality of care.
©Jinying Chen, Jiaping Zheng, Hong Yu. Originally published in JMIR Medical Informatics (http://medinform.jmir.org), 30.11.2016.
Keywords: electronic health records; information extraction; learning to rank; natural language processing; supervised learning
Conflict of interest statement
Conflicts of Interest: None declared.
References
- Proc Conf. 2016 Jun;2016:473-482 - PubMed
- J Biomed Inform. 2006 Dec;39(6):668-79 - PubMed
- Psychooncology. 2003 Sep;12(6):557-66 - PubMed
- J Med Internet Res. 2013 Mar 27;15(3):e65 - PubMed
- AMIA Annu Symp Proc. 2005;:859-63 - PubMed
- Health Serv Res. 2014 Feb;49(1 Pt 2):325-46 - PubMed
- Ann Intern Med. 2010 Jul 20;153(2):121-5 - PubMed
- Phys Sportsmed. 2014 Nov;42(4):125-30 - PubMed
- Methods Inf Med. 2002;41(4):289-98 - PubMed
- Respir Care. 2008 Oct;53(10):1310-5 - PubMed
- AMIA Annu Symp Proc. 2003;:674-8 - PubMed
- Br Med J (Clin Res Ed). 1986 Mar 1;292(6520):596-8 - PubMed
- AMIA Jt Summits Transl Sci Proc. 2015 Mar 25;2015:209-15 - PubMed
- Ann Intern Med. 2012 Oct 2;157(7):461-70 - PubMed
- J Med Internet Res. 2013 Aug 26;15(8):e168 - PubMed
- BMJ Qual Saf. 2016 May 6;:null - PubMed
- Biomed Res Int. 2014;2014:240403 - PubMed
- J Am Med Inform Assoc. 2010 May-Jun;17(3):229-36 - PubMed
- J Med Internet Res. 2015 Jun 23;17(6):e148 - PubMed
- Med Care. 2013 Mar;51(3 Suppl 1):S52-6 - PubMed
- CA Cancer J Clin. 1998 May-Jun;48(3):151-62 - PubMed
- J Bioinform Comput Biol. 2015 Jun;13(3):1541001 - PubMed
- Proc AMIA Symp. 1999;:107-11 - PubMed
- Stud Health Technol Inform. 2013;192:714-8 - PubMed
- J Health Commun. 2010;15 Suppl 2:183-96 - PubMed
- J Biomed Inform. 2015 Oct;57:333-49 - PubMed
- Stud Health Technol Inform. 2001;84(Pt 1):399-403 - PubMed
- AMIA Annu Symp Proc. 2007 Oct 11;:399-403 - PubMed
- N Engl J Med. 2009 Mar 12;360(11):1057-60 - PubMed
- Am J Kidney Dis. 2015 Jun;65(6):842-50 - PubMed
- J Am Med Inform Assoc. 2006 Jan-Feb;13(1):24-9 - PubMed
- J Med Internet Res. 2015 Dec 03;17(12):e275 - PubMed
- Am J Emerg Med. 2000 Nov;18(7):764-6 - PubMed
- AMIA Annu Symp Proc. 2007 Oct 11;:846-50 - PubMed
- AMIA Annu Symp Proc. 2010 Nov 13;2010:366-70 - PubMed
- Health Bull (Edinb). 1992 Mar;50(2):143-50 - PubMed
- Br J Gen Pract. 2004 Jan;54(498):38-43 - PubMed
- Stud Health Technol Inform. 2007;129(Pt 2):1117-21 - PubMed
- J Med Internet Res. 2001 Jul-Sep;3(3):E24 - PubMed
- J Am Med Inform Assoc. 2010 Sep-Oct;17(5):507-13 - PubMed
- J Am Med Inform Assoc. 2008 Jul-Aug;15(4):496-505 - PubMed
Publication Types
Grant support