Automated Detection of Periprosthetic Joint Infections and Data Elements Using Natural Language Processing.
artificial intelligence
electronic health records
natural language processing
periprosthetic joint infection
total joint arthroplasty
Journal
The Journal of arthroplasty
ISSN: 1532-8406
Titre abrégé: J Arthroplasty
Pays: United States
ID NLM: 8703515
Informations de publication
Date de publication:
02 2021
02 2021
Historique:
received:
09
04
2020
revised:
27
07
2020
accepted:
29
07
2020
pubmed:
29
8
2020
medline:
30
4
2021
entrez:
29
8
2020
Statut:
ppublish
Résumé
Periprosthetic joint infection (PJI) data elements are contained in both structured and unstructured documents in electronic health records and require manual data collection. The goal of this study is to develop a natural language processing (NLP) algorithm to replicate manual chart review for PJI data elements. PJI was identified among all total joint arthroplasty (TJA) procedures performed at a single academic institution between 2000 and 2017. Data elements that comprise the Musculoskeletal Infection Society (MSIS) criteria were manually extracted and used as the gold standard for validation. A training sample of 1208 TJA surgeries (170 PJI cases) was randomly selected to develop the prototype NLP algorithms and an additional 1179 surgeries (150 PJI cases) were randomly selected as the test sample. The algorithms were applied to all consultation notes, operative notes, pathology reports, and microbiology reports to predict the correct status of PJI based on MSIS criteria. The algorithm, which identified patients with PJI based on MSIS criteria, achieved an f1-score (harmonic mean of precision and recall) of 0.911. Algorithm performance in extracting the presence of sinus tract, purulence, pathologic documentation of inflammation, and growth of cultured organisms from the involved TJA achieved f1-scores that ranged from 0.771 to 0.982, sensitivity that ranged from 0.730 to 1.000, and specificity that ranged from 0.947 to 1.000. NLP-enabled algorithms have the potential to automate data collection for PJI diagnostic elements, which could directly improve patient care and augment cohort surveillance and research efforts. Further validation is needed in other hospital settings. Level III, Diagnostic.
Sections du résumé
BACKGROUND
Periprosthetic joint infection (PJI) data elements are contained in both structured and unstructured documents in electronic health records and require manual data collection. The goal of this study is to develop a natural language processing (NLP) algorithm to replicate manual chart review for PJI data elements.
METHODS
PJI was identified among all total joint arthroplasty (TJA) procedures performed at a single academic institution between 2000 and 2017. Data elements that comprise the Musculoskeletal Infection Society (MSIS) criteria were manually extracted and used as the gold standard for validation. A training sample of 1208 TJA surgeries (170 PJI cases) was randomly selected to develop the prototype NLP algorithms and an additional 1179 surgeries (150 PJI cases) were randomly selected as the test sample. The algorithms were applied to all consultation notes, operative notes, pathology reports, and microbiology reports to predict the correct status of PJI based on MSIS criteria.
RESULTS
The algorithm, which identified patients with PJI based on MSIS criteria, achieved an f1-score (harmonic mean of precision and recall) of 0.911. Algorithm performance in extracting the presence of sinus tract, purulence, pathologic documentation of inflammation, and growth of cultured organisms from the involved TJA achieved f1-scores that ranged from 0.771 to 0.982, sensitivity that ranged from 0.730 to 1.000, and specificity that ranged from 0.947 to 1.000.
CONCLUSION
NLP-enabled algorithms have the potential to automate data collection for PJI diagnostic elements, which could directly improve patient care and augment cohort surveillance and research efforts. Further validation is needed in other hospital settings.
LEVEL OF EVIDENCE
Level III, Diagnostic.
Identifiants
pubmed: 32854996
pii: S0883-5403(20)30869-X
doi: 10.1016/j.arth.2020.07.076
pmc: PMC7855617
mid: NIHMS1623200
pii:
doi:
Types de publication
Journal Article
Research Support, N.I.H., Extramural
Langues
eng
Sous-ensembles de citation
IM
Pagination
688-692Subventions
Organisme : NIAMS NIH HHS
ID : P30 AR076312
Pays : United States
Organisme : NIAMS NIH HHS
ID : R01 AR073147
Pays : United States
Informations de copyright
Copyright © 2020 Elsevier Inc. All rights reserved.
Références
J Arthroplasty. 2014 Jul;29(7):1331
pubmed: 24768547
NPJ Digit Med. 2019 Dec 17;2:130
pubmed: 31872069
J Surg Res. 2017 Mar;209:168-173
pubmed: 28032554
J Bone Joint Surg Am. 2019 Nov 6;101(21):1931-1938
pubmed: 31567670
Med Care. 2013 Jun;51(6):509-16
pubmed: 23673394
J Healthc Inform Res. 2019 Sep;3(3):267-282
pubmed: 31728432
J Arthroplasty. 2012 Sep;27(8 Suppl):61-5.e1
pubmed: 22554729
JAMA. 2011 Aug 24;306(8):848-55
pubmed: 21862746
Clin Infect Dis. 2013 Jan;56(1):e1-e25
pubmed: 23223583
Clin Orthop Relat Res. 2018 Feb;476(2):420-426
pubmed: 29389795
J Bone Joint Surg Am. 2019 Dec 18;101(24):2167-2174
pubmed: 31596819
AMIA Annu Symp Proc. 2018 Apr 16;2017:515-524
pubmed: 29854116
J Arthroplasty. 2019 Oct;34(10):2216-2219
pubmed: 31416741