An evaluation of medical knowledge contained in Wikipedia and its use in the LOINC database

From WikiLit
Jump to: navigation, search
Publication (help)
An evaluation of medical knowledge contained in Wikipedia and its use in the LOINC database
Authors: Jeff Friedlin, Clement J McDonald [edit item]
Citation: Journal of the American Medical Informatics Association 17 (3): 283-287. 2010 May.
Publication type: Journal article
Peer-reviewed: Yes
Database(s):
DOI: 10.1136/jamia.2009.001180.
Google Scholar cites: Citations
Link(s): Paper link
Added by Wikilit team: Added on initial load
Search
Article: Google Scholar BASE PubMed
Other scholarly wikis: AcaWiki Brede Wiki WikiPapers
Web search: Bing Google Yahoo!Google PDF
Other:
Services
Format: BibTeX
An evaluation of medical knowledge contained in Wikipedia and its use in the LOINC database is a publication by Jeff Friedlin, Clement J McDonald.


[edit] Abstract

The logical observation identifiers names and codes {(LOINC)} database contains 55 000 terms consisting of more atomic components called parts. {LOINC} carries more than 18 000 distinct parts. It is necessary to have definitions/descriptions for each of these parts to assist users in mapping local laboratory codes to {LOINC.} It is believed that much of this information can be obtained from the internet; the first effort was with Wikipedia. This project focused on 1705 laboratory analytes (the first part in the {LOINC} laboratory name). Of the 1705 parts queried, 1314 matching articles were found in Wikipedia. Of these, 1299 (98.9\%) were perfect matches that exactly described the {LOINC} part, 15 (1.14\%) were partial matches (the description in Wikipedia was related to the {LOINC} part, but did not describe it fully), and 102 (7.76\%) were mis-matches. The current release of {RELMA} and {LOINC} include Wikipedia descriptions of {LOINC} parts obtained as a direct result of this project.

[edit] Research questions

"The purpose of this study was twofold. First, we wished to evaluate the degree of medical knowledge contained in the online encyclopedia Wikipedia and the feasibility of using that knowledge as a means of adding description information to a laboratory and clinical observations database (LOINC). Second, we desired to test our software’s ability to automatically extract relevant information from Wikipedia based on queries generated from part names taken from the LOINC part database."

Research details

Topics: Other information retrieval topics [edit item]
Domains: Computer science, Health [edit item]
Theory type: Design and action [edit item]
Wikipedia coverage: Sample data [edit item]
Theories: "Undetermined" [edit item]
Research design: Design science, Experiment [edit item]
Data source: Archival records, Experiment responses, Wikipedia pages [edit item]
Collected data time dimension: Cross-sectional [edit item]
Unit of analysis: Subject [edit item]
Wikipedia data extraction: Live Wikipedia [edit item]
Wikipedia page type: Article [edit item]
Wikipedia language: English [edit item]

[edit] Conclusion

"We conclude that Wikipedia contains a surprisingly large amount of scientific and medical data and could effectively be used as an initial knowledge base for specific medical informatics and research projects. The software we developed to automate the matching of LOINC part names to Wikipedia articles performed satisfactorily with high sensitivity and moderate specificity. The current release of RELMA and LOINC include descriptions of LOINC parts obtained from Wikipedia as a direct result of this project."

[edit] Comments

"Wikipedia pages + LOINC database

"Research design" should also include "Design science". "Experiment" is perhaps also ok."


Further notes[edit]

Facts about "An evaluation of medical knowledge contained in Wikipedia and its use in the LOINC database"RDF feed
AbstractThe logical observation identifiers names The logical observation identifiers names and codes {(LOINC)} database contains 55 000 terms consisting of more atomic components called parts. {LOINC} carries more than 18 000 distinct parts. It is necessary to have definitions/descriptions for each of these parts to assist users in mapping local laboratory codes to {LOINC.} It is believed that much of this information can be obtained from the internet; the first effort was with Wikipedia. This project focused on 1705 laboratory analytes (the first part in the {LOINC} laboratory name). Of the 1705 parts queried, 1314 matching articles were found in Wikipedia. Of these, 1299 (98.9\%) were perfect matches that exactly described the {LOINC} part, 15 (1.14\%) were partial matches (the description in Wikipedia was related to the {LOINC} part, but did not describe it fully), and 102 (7.76\%) were mis-matches. The current release of {RELMA} and {LOINC} include Wikipedia descriptions of {LOINC} parts obtained as a direct result of this project.tained as a direct result of this project.
Added by wikilit teamAdded on initial load +
Collected data time dimensionCross-sectional +
CommentsWikipedia pages +

LOINC database

"Research design" should also include "Design science". "Experiment" is perhaps also ok.
ConclusionWe conclude that Wikipedia contains a surpWe conclude that Wikipedia contains a surprisingly large

amount of scientific and medical data and could effectively be used as an initial knowledge base for specific medical informatics and research projects. The software we developed to automate the matching of LOINC part names to Wikipedia articles performed satisfactorily with high sensitivity and moderate specificity. The current release of RELMA and LOINC include descriptions of LOINC parts obtained from Wikipedia as a direct result of this project.ipedia as a direct

result of this project.
Data sourceArchival records +, Experiment responses + and Wikipedia pages +
Doi10.1136/jamia.2009.001180 +
Google scholar urlhttp://scholar.google.com/scholar?ie=UTF-8&q=%22An%2Bevaluation%2Bof%2Bmedical%2Bknowledge%2Bcontained%2Bin%2BWikipedia%2Band%2Bits%2Buse%2Bin%2Bthe%2BLOINC%2Bdatabase%22 +
Has authorJeff Friedlin + and Clement J McDonald +
Has domainComputer science + and Health +
Has topicOther information retrieval topics +
Issue3 +
MonthMay +
Pages283-287 +
Peer reviewedYes +
Publication typeJournal article +
Published inJournal of the American Medical Informatics Association +
Research designDesign science + and Experiment +
Research questionsThe purpose of this study was twofold. FirThe purpose of this study was twofold. First, we wished to

evaluate the degree of medical knowledge contained in the online encyclopedia Wikipedia and the feasibility of using that knowledge as a means of adding description information to a laboratory and clinical observations database (LOINC). Second, we desired to test our software’s ability to automatically extract relevant information from Wikipedia based on queries generated

from part names taken from the LOINC part database.
names taken from the LOINC part database.
Revid10,659 +
TheoriesUndetermined
Theory typeDesign and action +
TitleAn evaluation of medical knowledge contained in Wikipedia and its use in the LOINC database
Unit of analysisSubject +
Urlhttp://www.ncbi.nlm.nih.gov/pubmed/20442145 +
Volume17 +
Wikipedia coverageSample data +
Wikipedia data extractionLive Wikipedia +
Wikipedia languageEnglish +
Wikipedia page typeArticle +
Year2010 +