Treffer: Polar labeling: silver standard algorithm for training disease classifiers.

Title:
Polar labeling: silver standard algorithm for training disease classifiers.
Authors:
Wagholikar KB; Laboratory of Computer Science, Massachusetts General Hospital, Boston, MA 02114, USA., Estiri H; Laboratory of Computer Science, Massachusetts General Hospital, Boston, MA 02114, USA., Murphy M; Partners Healthcare, Somerville, MA 02145, USA., Murphy SN; Laboratory of Computer Science, Massachusetts General Hospital, Boston, MA 02114, USA.
Source:
Bioinformatics (Oxford, England) [Bioinformatics] 2020 May 01; Vol. 36 (10), pp. 3200-3206.
Publication Type:
Journal Article; Research Support, N.I.H., Extramural
Language:
English
Journal Info:
Publisher: Oxford University Press Country of Publication: England NLM ID: 9808944 Publication Model: Print Cited Medium: Internet ISSN: 1367-4811 (Electronic) Linking ISSN: 13674803 NLM ISO Abbreviation: Bioinformatics Subsets: MEDLINE
Imprint Name(s):
Original Publication: Oxford : Oxford University Press, c1998-
MeSH Terms:
Algorithms* , Machine Learning*, Color ; Humans
References:
J Am Med Inform Assoc. 2015 Sep;22(5):993-1000. (PMID: 25929596)
J Am Med Inform Assoc. 2017 Jan;24(1):162-171. (PMID: 27497800)
J Am Med Inform Assoc. 2017 Apr 1;24(e1):e143-e149. (PMID: 27632993)
J Am Med Inform Assoc. 2014 Mar-Apr;21(2):221-30. (PMID: 24201027)
J Am Med Inform Assoc. 2013 Jan 1;20(1):117-21. (PMID: 22955496)
J Am Med Inform Assoc. 2013 Dec;20(e2):e226-31. (PMID: 23956018)
Mol Oncol. 2015 May;9(5):1018-24. (PMID: 25248998)
J Biomed Inform. 2014 Dec;52:199-211. (PMID: 25038555)
J Diabetes Sci Technol. 2017 Jul;11(4):791-799. (PMID: 27932531)
J Am Med Inform Assoc. 2018 Jan 1;25(1):54-60. (PMID: 29126253)
AMIA Annu Symp Proc. 2011;2011:189-96. (PMID: 22195070)
J Am Med Inform Assoc. 2013 Dec;20(e2):e253-9. (PMID: 23851443)
AMIA Annu Symp Proc. 2008 Nov 06;:1170. (PMID: 18999043)
Evid Based Ment Health. 2017 Aug;20(3):83-87. (PMID: 28739578)
J Am Med Inform Assoc. 2017 Sep 1;24(5):996-1001. (PMID: 28340241)
J Am Med Inform Assoc. 2015 Nov;22(6):1251-60. (PMID: 26224336)
AMIA Annu Symp Proc. 2006;:1044. (PMID: 17238663)
Artif Intell Med. 2007 Nov;41(3):251-62. (PMID: 17707617)
J Am Med Inform Assoc. 2016 Nov;23(6):1166-1173. (PMID: 27174893)
Grant Information:
R01 HG009174 United States HG NHGRI NIH HHS; R00 LM011575 United States LM NLM NIH HHS
Entry Date(s):
Date Created: 20200213 Date Completed: 20201029 Latest Revision: 20201029
Update Code:
20250114
PubMed Central ID:
PMC7214041
DOI:
10.1093/bioinformatics/btaa088
PMID:
32049335
Database:
MEDLINE

Weitere Informationen

Motivation: Expert-labeled data are essential to train phenotyping algorithms for cohort identification. However expert labeling is time and labor intensive, and the costs remain prohibitive for scaling phenotyping to wider use-cases.
Results: We present an approach referred to as polar labeling (PL), to create silver standard for training machine learning (ML) for disease classification. We test the hypothesis that ML models trained on the silver standard created by applying PL on unlabeled patient records, are comparable in performance to the ML models trained on gold standard, created by clinical experts through manual review of patient records. We perform experimental validation using health records of 38 023 patients spanning six diseases. Our results demonstrate the superior performance of the proposed approach.
Availability and Implementation: We provide a Python implementation of the algorithm and the Python code developed for this study on Github.
Supplementary Information: Supplementary data are available at Bioinformatics online.
(© The Author(s) 2020. Published by Oxford University Press.)