Treffer: Using Natural Language Processing of Free-Text Radiology Reports to Identify Type 1 Modic Endplate Changes.

Title:
Using Natural Language Processing of Free-Text Radiology Reports to Identify Type 1 Modic Endplate Changes.
Authors:
Huhdanpaa HT; Radia, Inc., Lynwood, WA, USA., Tan WK; Department of Biostatistics, University of Washington, Seattle, WA, USA.; Center for Biomedical Statistics, University of Washington, Seattle, WA, USA., Rundell SD; Department of Rehabilitation Medicine, University of Washington, Seattle, WA, USA.; Comparative Effectiveness, Cost and Outcomes Research Center, University of Washington, Seattle, WA, USA., Suri P; Department of Rehabilitation Medicine, University of Washington, Seattle, WA, USA.; Comparative Effectiveness, Cost and Outcomes Research Center, University of Washington, Seattle, WA, USA.; Division of Rehabilitation Care Services, Seattle Epidemiologic Research and Information Center, VA Puget Sound Health Care System, Seattle, WA, USA., Chokshi FH; Department of Radiology and Imaging Sciences, Emory University School of Medicine, Atlanta, GA, USA., Comstock BA; Department of Biostatistics, University of Washington, Seattle, WA, USA.; Center for Biomedical Statistics, University of Washington, Seattle, WA, USA., Heagerty PJ; Department of Biostatistics, University of Washington, Seattle, WA, USA.; Center for Biomedical Statistics, University of Washington, Seattle, WA, USA., James KT; Comparative Effectiveness, Cost and Outcomes Research Center, University of Washington, Seattle, WA, USA.; Department of Radiology, University of Washington, Box 359728, 325 Ninth Ave., Seattle, WA, 98104-2499, USA., Avins AL; Division of Research, Kaiser Permanente Northern California, Oakland, CA, USA., Nedeljkovic SS; Department of Anesthesiology, Perioperative and Pain Medicine, Harvard Vanguard Medical Associates, Brigham and Women's Hospital and Spine Unit, Boston, MA, USA., Nerenz DR; Henry Ford Hospital, Neuroscience Institute, Detroit, MI, USA., Kallmes DF; Department of Radiology, Mayo Clinic, Rochester, MN, USA., Luetmer PH; Department of Radiology, Mayo Clinic, Rochester, MN, USA., Sherman KJ; Kaiser Permanente of Washington Research Institute, Seattle, WA, USA., Organ NL; Department of Biostatistics, University of Washington, Seattle, WA, USA.; Center for Biomedical Statistics, University of Washington, Seattle, WA, USA., Griffith B; Department of Radiology, Henry Ford Hospital, Detroit, MI, USA., Langlotz CP; Department of Radiology, Stanford University, Palo Alto, CA, USA., Carrell D; Kaiser Permanente of Washington Research Institute, Seattle, WA, USA., Hassanpour S; Department of Biomedical Data Science, Dartmouth College, Lebanon, NH, USA., Jarvik JG; Comparative Effectiveness, Cost and Outcomes Research Center, University of Washington, Seattle, WA, USA. jarvikj@uw.edu.; Department of Radiology, University of Washington, Box 359728, 325 Ninth Ave., Seattle, WA, 98104-2499, USA. jarvikj@uw.edu.; Department of Neurological Surgery, University of Washington, Seattle, WA, USA. jarvikj@uw.edu.; Department of Health Services, University of Washington, Seattle, WA, USA. jarvikj@uw.edu.
Source:
Journal of digital imaging [J Digit Imaging] 2018 Feb; Vol. 31 (1), pp. 84-90.
Publication Type:
Journal Article; Research Support, N.I.H., Extramural; Research Support, U.S. Gov't, P.H.S.
Language:
English
Journal Info:
Publisher: Springer Country of Publication: United States NLM ID: 9100529 Publication Model: Print Cited Medium: Internet ISSN: 1618-727X (Electronic) Linking ISSN: 08971889 NLM ISO Abbreviation: J Digit Imaging Subsets: MEDLINE
Imprint Name(s):
Publication: <2008-2023>: New York : Springer
Original Publication: Philadelphia, PA : W.B. Saunders, c1988-
MeSH Terms:
Natural Language Processing* , Research Report*, Low Back Pain/*pathology , Lumbar Vertebrae/*diagnostic imaging , Lumbar Vertebrae/*pathology , Magnetic Resonance Imaging/*methods, Humans ; Prospective Studies ; Radiology ; Reproducibility of Results ; Sensitivity and Specificity
References:
Contemp Clin Trials. 2015 Nov;45(Pt B):157-163. (PMID: 26493088)
J Biomed Inform. 2009 Apr;42(2):377-81. (PMID: 18929686)
Eur Spine J. 2008 Nov;17(11):1407-22. (PMID: 18787845)
Radiology. 2009 Oct;253(1):23-5. (PMID: 19789252)
Radiology. 1988 Jan;166(1 Pt 1):193-9. (PMID: 3336678)
Spine J. 2014 Nov 1;14(11):2525-45. (PMID: 24768732)
Science. 2015 Jul 17;349(6245):261-6. (PMID: 26185244)
J Digit Imaging. 2010 Apr;23(2):119-32. (PMID: 19484309)
J Digit Imaging. 2012 Feb;25(1):30-6. (PMID: 22038514)
Radiology. 2016 May;279(2):329-43. (PMID: 27089187)
BMC Musculoskelet Disord. 2014 Apr 23;15:134. (PMID: 24755158)
J Biomed Inform. 2009 Oct;42(5):839-51. (PMID: 19435614)
Biometrics. 1977 Mar;33(1):159-74. (PMID: 843571)
J Biomed Inform. 2001 Oct;34(5):301-10. (PMID: 12123149)
Radiographics. 2016 Jan-Feb;36(1):176-91. (PMID: 26761536)
PLoS One. 2013 Jul 09;8(7):e67863. (PMID: 23874456)
J Am Coll Radiol. 2009 Dec;6(12 ):851-60. (PMID: 19945040)
Grant Information:
R01 HS022972 United States HS AHRQ HHS; UH3 AR066795 United States AR NIAMS NIH HHS; 1UH2AT007766-01 United States NH NIH HHS; 4UH3AR066795-02 United States NH NIH HHS
Contributed Indexing:
Keywords: Lumbar spine imaging; Modic classification; Natural language processing; Radiology reporting
Entry Date(s):
Date Created: 20170816 Date Completed: 20190501 Latest Revision: 20190501
Update Code:
20250114
PubMed Central ID:
PMC5788819
DOI:
10.1007/s10278-017-0013-3
PMID:
28808792
Database:
MEDLINE

Weitere Informationen

Electronic medical record (EMR) systems provide easy access to radiology reports and offer great potential to support quality improvement efforts and clinical research. Harnessing the full potential of the EMR requires scalable approaches such as natural language processing (NLP) to convert text into variables used for evaluation or analysis. Our goal was to determine the feasibility of using NLP to identify patients with Type 1 Modic endplate changes using clinical reports of magnetic resonance (MR) imaging examinations of the spine. Identifying patients with Type 1 Modic change who may be eligible for clinical trials is important as these findings may be important targets for intervention. Four annotators identified all reports that contained Type 1 Modic change, using N = 458 randomly selected lumbar spine MR reports. We then implemented a rule-based NLP algorithm in Java using regular expressions. The prevalence of Type 1 Modic change in the annotated dataset was 10%. Results were recall (sensitivity) 35/50 = 0.70 (95% confidence interval (C.I.) 0.52-0.82), specificity 404/408 = 0.99 (0.97-1.0), precision (positive predictive value) 35/39 = 0.90 (0.75-0.97), negative predictive value 404/419 = 0.96 (0.94-0.98), and F1-score 0.79 (0.43-1.0). Our evaluation shows the efficacy of rule-based NLP approach for identifying patients with Type 1 Modic change if the emphasis is on identifying only relevant cases with low concern regarding false negatives. As expected, our results show that specificity is higher than recall. This is due to the inherent difficulty of eliciting all possible keywords given the enormous variability of lumbar spine reporting, which decreases recall, while availability of good negation algorithms improves specificity.