| Autor: |
Koller W; Department for Hospital Epidemiology and Infection Control, Vienna General Hospital and Medical University of Vienna, Vienna, Austria., Kleinoscheg G; Medexter Healthcare GmbH, Vienna, Austria., Willinger B; Division of Clinical Microbiology, Vienna General Hospital and Medical University of Vienna, Vienna, Austria., Rappelsberger A; Section for Artificial Intelligence and Decision Support, Medical University of Vienna, Vienna, Austria., Adlassnig KP; Medexter Healthcare GmbH, Vienna, Austria.; Section for Artificial Intelligence and Decision Support, Medical University of Vienna, Vienna, Austria. |
| Jazyk: |
angličtina |
| Zdroj: |
Studies in health technology and informatics [Stud Health Technol Inform] 2019 Aug 21; Vol. 264, pp. 1243-1247. |
| DOI: |
10.3233/SHTI190425 |
| Abstrakt: |
In the present study, we intended to solve identification problems in analyzing the results of microbiology by proactive man-machine interaction. We modified the analytics software MOMO so that it flags laboratory results containing textual elements unknown to the thesaurus, and a human expert assigns the elements to the respective existing thesaurus elements or creates new ones. In 773,309 laboratory results, roughly 2.6% contained unassigned elements and would have been ignored in thesaurus-based analyses for purposes other than simply reporting microbiological findings to physicians. In current use, the thesaurus is kept up to date with synonyms, syntactic deviations, misspellings, and entries not contained earlier, with man-machine interaction of 2-3 hours per week. This approach helps to accommodate both up-to-date clinical reporting for immediate patient care as well as up-to-date queries for infection surveillance and epidemiology, outbreak management, quality control and benchmarking, and antimicrobial stewardship. |
| Databáze: |
MEDLINE |
| Externí odkaz: |
|
