Detection of Blood Culture Bacterial Contamination using Natural Language Processing
作者： Fern FitzHenry , Robert S. Dittus , Harvey J. Murff , Peter L. Elkin , Michael E. Matheny
DOI:
关键词: Detection performance 、 Predictive value 、 Data mining 、 Blood culture 、 Sensitivity (control systems) 、 Natural language processing 、 Contamination 、 Artificial intelligence 、 Test data 、 Medicine 、 False Negative Reactions 、 Data extraction
摘要: Microbiology results are reported in semi-structured formats and have a high content of useful patient information. We developed validated hybrid regular expression natural language processing solution for blood culture microbiology reports. Multi-center Veterans Affairs training testing data sets were randomly extracted manually reviewed to determine the sensitivity as well contamination results. The tool was iteratively both outcomes using dataset, then evaluated on test dataset antibiotic susceptibility extraction detection performance. Our algorithm had 84.8% positive predictive value 96.0% mapping antibiotics bacteria with appropriate findings data. bacterial 83.3% 81.8%.
researchgate.net 参考文章(11)
Dietlind L. Wahner-Roedler, Peter L. Elkin, Steven H. Brown, Brent A. Bauer, Coverage of oncology drug indication concepts and compositional semantics by SNOMED-CT. american medical informatics association annual symposium. ,vol. 2003, pp. 115- 119 ,(2003)
Shaun J. Grannis, J. Marc Overhage, Jeff Friedlin, Using natural language processing to improve accuracy of automated notifiable disease reporting. american medical informatics association annual symposium. ,vol. 2008, pp. 207- 211 ,(2008)
Christopher G. Chute, Peter L. Elkin, Kent R. Bailey, A randomized controlled trial of automated term composition. american medical informatics association annual symposium. pp. 765- 769 ,(1998)
Marc J. Overhage, Jeffrey Suico, Clement J. McDonald, Electronic laboratory reporting: barriers, solutions and findings. Journal of Public Health Management and Practice. ,vol. 7, pp. 60- 66 ,(2001) , 10.1097/00124784-200107060-00007
Manuel D. J. Arbo, Influence of Blood Culture Results on Antibiotic Choice in the Treatment of Bacteremia Archives of Internal Medicine. ,vol. 154, pp. 2641- 2645 ,(1994) , 10.1001/ARCHINTE.1994.00420230024004
P.L Elkin, S.H Brown, Automated enhancement of description logic-defined terminologies to facilitate mapping to ICD9-CM Journal of Biomedical Informatics. ,vol. 35, pp. 281- 288 ,(2002) , 10.1016/S1532-0464(03)00019-4
Dag Berild, Atefeh Mohseni, Lien My Diep, Mogens Jensenius, Signe Holta Ringertz, Adjustment of antibiotic treatment according to the results of blood cultures leads to decreased antibiotic use and costs Journal of Antimicrobial Chemotherapy. ,vol. 57, pp. 326- 330 ,(2006) , 10.1093/JAC/DKI463
Joseph A Tworek, Molly K Walsh, Leonas G Bekeris, Paul N Valenstein, Trends in Blood Culture Contamination: A College of American Pathologists Q-Tracks Study of 356 Institutions Archives of Pathology & Laboratory Medicine. ,vol. 129, pp. 1222- 1225 ,(2005) , 10.1043/1543-2165(2005)129[1222:TIBCCA]2.0.CO;2
Steven H. Brown, Theodore Speroff, Elliot M. Fielstein, Brent A. Bauer, Dietlind L. Wahner-Roedler, Robert Greevy, Peter L. Elkin, eQuality: Electronic Quality Assessment From Narrative Clinical Reports Mayo Clinic Proceedings. ,vol. 81, pp. 1472- 1481 ,(2006) , 10.4065/81.11.1472
Keith F. Woeltje, Anne M. Butler, Ashleigh J. Goris, Nhial T. Tutlam, Joshua A. Doherty, M. Brandon Westover, Vicky Ferris, Thomas C. Bailey, Automated surveillance for central line-associated bloodstream infection in intensive care units. Infection Control and Hospital Epidemiology. ,vol. 29, pp. 842- 846 ,(2008) , 10.1086/590261