Automated information extraction from free‐text medical documents for stroke key performance indicators: a pilot study
Stephen Bacchi, Samuel Gluck, Simon A. Koblar, Jim Jannes, Timothy Kleinig
Abstract
Automated information extraction might be able to assist with the collection of stroke key performance indicators (KPI). The feasibility of using natural language processing for classification-based KPI and datetime field extraction was assessed. Using free-text discharge summaries, random forest models achieved high levels of performance in classification tasks (area under the receiver operator curve 0.95-1.00). The datetime field extraction method was successful in 29 of 43 (67.4%) cases. Further studies are indicated.
Topics & Concepts
MedicineKey (lock)Receiver operating characteristicRandom forestField (mathematics)Information extractionExtraction (chemistry)Natural language processingPerformance indicatorInformation retrievalArtificial intelligenceData miningComputer scienceInternal medicineManagementChromatographyMathematicsComputer securityEconomicsChemistryPure mathematicsBiomedical Text Mining and OntologiesTopic ModelingText Readability and Simplification