Abstract
Data mining in electronic medical records may facilitate clinical research, but much of the structured data may be miscoded, incomplete, or non-specific. The exploitation of narrative data using natural language processing may help, although nesting, varying granularity, and repetition remain challenges. In a study of community-acquired pneumonia using electronic records, these issues led to poor classification. Limiting queries to accurate, complete records led to vastly reduced, possibly biased samples. We exploited knowledge latent in the electronic records to improve classification. A similarity metric was used to cluster cases. We defined discordance as the degree to which cases within a cluster give different answers for some query that addresses a classification task of interest. Cases with higher discordance are more likely to be incorrectly classified, and can be reviewed manually to adjust the classification, improve the query, or estimate the likely accuracy of the query. In a study of pneumonia--in which the ICD9-CM coding was found to be very poor--the discordance measure was statistically significantly correlated with classification correctness (.45; 95% CI .15-.62).
References
Apr 1, 1991·Statistics in Medicine·C Safran
Feb 11, 1988·The New England Journal of Medicine·D C HsiaR P Kusserow
Jan 1, 1987·Journal of Chronic Diseases·M E CharlsonC R MacKenzie
Feb 1, 1984·Archives of Internal Medicine·J M Harris
Mar 1, 1994·Journal of the American Medical Informatics Association : JAMIA·N SagerL J Tick
Sep 1, 1994·Journal of Clinical Epidemiology·D J MalenkaJ E Wennberg
Sep 1, 1994·Journal of the American Medical Informatics Association : JAMIA·D F Sittig
Oct 15, 1993·Annals of Internal Medicine·J G JollisD B Mark
Apr 1, 1993·Medical Informatics = Médecine Et Informatique·E KrusinskaL Franzén
Aug 1, 1996·Journal of Medical Systems·N Sakamoto
Jan 23, 1997·The New England Journal of Medicine·M J FineW N Kapoor
Jul 1, 1997·Journal of the American Medical Informatics Association : JAMIA·W M TierneyX H Zhou
Feb 12, 1998·Annals of Internal Medicine·L I Iezzoni
Nov 24, 1998·Journal of the American Medical Informatics Association : JAMIA·P M Nadkarni, C Brandt
Sep 24, 1999·Journal of the American Medical Informatics Association : JAMIA·D B AronowW B Croft
Jan 21, 2000·Journal of the American Medical Informatics Association : JAMIA·D Aronsky, P J Haug
Apr 25, 2000·Computers and Biomedical Research, an International Journal·J S ElkinsG Hripcsak
Jul 11, 2000·Journal of the American Medical Informatics Association : JAMIA·P M NadkarniL Marenco
Nov 4, 2000·Journal of the American Medical Informatics Association : JAMIA·M FiszmanP J Haug
May 19, 2001·BMJ : British Medical Journal·D M CoulterI R Edwards
Nov 1, 2001·Journal of the American Medical Informatics Association : JAMIA·P G MutalikP M Nadkarni
Jul 2, 2002·Radiology·George HripcsakCarol Friedman
Mar 15, 2003·Artificial Intelligence in Medicine·Mia K MarkeyCarey E Floyd
Apr 2, 2003·Journal of the American Medical Informatics Association : JAMIA·Adam B Wilcox, George Hripcsak
May 4, 2004·Journal of Biomedical Informatics·Jorge E Caviedes, James J Cimino
Jun 10, 2004·Journal of the American Medical Informatics Association : JAMIA·Carol FriedmanGeorge Hripcsak
Dec 14, 2004·Medical Care·John W PeabodyPeter Glassman
Dec 25, 2004·Journal of Clinical Epidemiology·David B PreenJames B Semmens
Aug 9, 2005·Journal of Biomedical Informatics·Eneida A MendonçaCarol Friedman
Citations
Dec 29, 2012·PloS One·D J AlbersMichael Schmidt
Aug 24, 2013·PloS One·Sylvain DeLisleLeonard D'Avolio
Jun 8, 2011·Journal of Biomedical Discovery and Collaboration·George HripcsakGenevieve Melton
Sep 8, 2012·Journal of the American Medical Informatics Association : JAMIA·George Hripcsak, David J Albers
Nov 2, 2013·Journal of Biomedical Informatics·Vishal N Patel, David C Kaelber
Jan 22, 2013·The Journal of School Health·Ahmed H YoussefAghaWasantha P Jayawardene
Jan 15, 2014·Pharmacotherapy·Aisling R CaffreyKerry L Laplante
Jun 5, 2015·Journal of the American Medical Informatics Association : JAMIA·Mary Regina BolandNicholas P Tatonetti
Jul 22, 2017·Journal of Biomedical Informatics·Kory KreimeyerTaxiarchis Botsis