Using Artificial Intelligence With Natural Language Processing to Combine Electronic Health Record's Structured and Free Text Data to Identify Nonvalvular Atrial Fibrillation to Decrease Strokes and Death: Evaluation and Case-Control Study.
CHA2DS2-VASc
HAS-BLED
NVAF
afib
artificial intelligence
atrial fibrillation
bio-surveillance
bleed risk
natural language processing
stroke risk
Journal
Journal of medical Internet research
ISSN: 1438-8871
Titre abrégé: J Med Internet Res
Pays: Canada
ID NLM: 100959882
Informations de publication
Date de publication:
09 11 2021
09 11 2021
Historique:
received:
19
03
2021
accepted:
05
07
2021
revised:
05
06
2021
entrez:
9
11
2021
pubmed:
10
11
2021
medline:
16
11
2021
Statut:
epublish
Résumé
Nonvalvular atrial fibrillation (NVAF) affects almost 6 million Americans and is a major contributor to stroke but is significantly undiagnosed and undertreated despite explicit guidelines for oral anticoagulation. The aim of this study is to investigate whether the use of semisupervised natural language processing (NLP) of electronic health record's (EHR) free-text information combined with structured EHR data improves NVAF discovery and treatment and perhaps offers a method to prevent thousands of deaths and save billions of dollars. We abstracted 96,681 participants from the University of Buffalo faculty practice's EHR. NLP was used to index the notes and compare the ability to identify NVAF, congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, stroke or transient ischemic attack, vascular disease, age 65 to 74 years, sex category (CHA The structured-plus-unstructured method would have identified 3,976,056 additional true NVAF cases (P<.001) and improved sensitivity for CHA Artificial intelligence-informed bio-surveillance combining NLP of free-text information with structured EHR data improves data completeness, prevents thousands of strokes, and saves lives and funds. This method is applicable to many disorders with profound public health consequences.
Sections du résumé
BACKGROUND
Nonvalvular atrial fibrillation (NVAF) affects almost 6 million Americans and is a major contributor to stroke but is significantly undiagnosed and undertreated despite explicit guidelines for oral anticoagulation.
OBJECTIVE
The aim of this study is to investigate whether the use of semisupervised natural language processing (NLP) of electronic health record's (EHR) free-text information combined with structured EHR data improves NVAF discovery and treatment and perhaps offers a method to prevent thousands of deaths and save billions of dollars.
METHODS
We abstracted 96,681 participants from the University of Buffalo faculty practice's EHR. NLP was used to index the notes and compare the ability to identify NVAF, congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, stroke or transient ischemic attack, vascular disease, age 65 to 74 years, sex category (CHA
RESULTS
The structured-plus-unstructured method would have identified 3,976,056 additional true NVAF cases (P<.001) and improved sensitivity for CHA
CONCLUSIONS
Artificial intelligence-informed bio-surveillance combining NLP of free-text information with structured EHR data improves data completeness, prevents thousands of strokes, and saves lives and funds. This method is applicable to many disorders with profound public health consequences.
Identifiants
pubmed: 34751659
pii: v23i11e28946
doi: 10.2196/28946
pmc: PMC8663460
doi:
Substances chimiques
Anticoagulants
0
Types de publication
Journal Article
Research Support, N.I.H., Extramural
Langues
eng
Sous-ensembles de citation
IM
Pagination
e28946Subventions
Organisme : NLM NIH HHS
ID : T15 LM007056
Pays : United States
Organisme : NCATS NIH HHS
ID : UL1 TR001412
Pays : United States
Organisme : NIAAA NIH HHS
ID : R21 AA026954
Pays : United States
Organisme : NLM NIH HHS
ID : T15 LM012495
Pays : United States
Organisme : NIAAA NIH HHS
ID : R33 AA026954
Pays : United States
Organisme : NIGMS NIH HHS
ID : T32 GM099607
Pays : United States
Informations de copyright
©Peter L Elkin, Sarah Mullin, Jack Mardekian, Christopher Crowner, Sylvester Sakilay, Shyamashree Sinha, Gary Brady, Marcia Wright, Kimberly Nolen, JoAnn Trainer, Ross Koppel, Daniel Schlegel, Sashank Kaushik, Jane Zhao, Buer Song, Edwin Anand. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 09.11.2021.
Références
Appl Clin Inform. 2011 Dec 21;2(4):534-45
pubmed: 23616894
J Am Coll Cardiol. 2011 Jan 11;57(2):173-80
pubmed: 21111555
Circulation. 2012 Jan 3;125(1):e2-e220
pubmed: 22179539
Lancet. 2016 Aug 20;388(10046):806-17
pubmed: 27560276
Chest. 2010 Nov;138(5):1093-100
pubmed: 20299623
Appl Clin Inform. 2017 Apr 12;8(2):369-380
pubmed: 28401245
Med Care. 2017 Mar;55(3):244-251
pubmed: 27787351
Ann Intern Med. 2012 Jan 3;156(1 Pt 1):11-8
pubmed: 22213490
Eur Heart J. 2012 Nov;33(21):2719-47
pubmed: 22922413
PLoS One. 2016 Apr 29;11(4):e0154556
pubmed: 27128319
JAMA. 2001 Jun 13;285(22):2864-70
pubmed: 11401607
Med Care. 2005 May;43(5):480-5
pubmed: 15838413
Am J Health Syst Pharm. 2011 Jul 1;68(13):1239-44
pubmed: 21690430
Am J Cardiovasc Drugs. 2017 Dec;17(6):481-492
pubmed: 28795348
Biometrics. 2000 Jun;56(2):345-51
pubmed: 10877288
Per Med. 2008 May;5(3):279-284
pubmed: 29783488
Am Heart J. 2015 Jan;169(1):39-44.e2
pubmed: 25497246
J Biomed Inform. 2009 Oct;42(5):760-72
pubmed: 19683066
N Engl J Med. 2018 Apr 5;378(14):1271-1273
pubmed: 29617592
JAMA. 2011 Aug 24;306(8):848-55
pubmed: 21862746
Stroke. 1991 Aug;22(8):983-8
pubmed: 1866765
Chest. 2010 Feb;137(2):263-72
pubmed: 19762550
Lancet Public Health. 2018 Apr;3(4):e185-e193
pubmed: 29550372
Pharmacogenomics. 2010 Jul;11(7):989-1002
pubmed: 20602617
J Am Heart Assoc. 2016 Jun 13;5(6):
pubmed: 27412905
Circulation. 2019 Jul 9;140(2):e125-e151
pubmed: 30686041
Stud Health Technol Inform. 2017;235:276-280
pubmed: 28423797
J Am Coll Cardiol. 2014 Dec 2;64(21):e1-76
pubmed: 24685669