Automated information extraction from free‐text medical documents for stroke key performance indicators: a pilot study. Issue 2 (20th February 2022)

• Record Type: Journal Article
• Authors: Bacchi, Stephen; Gluck, Sam; Koblar, Simon; Jannes, Jim; Kleinig, Timothy
• Is Part Of: Internal medicine journal. Volume 52:Issue 2(2022)
• Page Start: 315
• Page End: 317
• Publication Date: 2022-02-20
• Subjects: natural language processing -- machine learning -- random forest -- key performance indicator; Medicine -- Periodicals; 616
• DOI: 10.1111/imj.15678
• Languages: English
• ISSNs: 1444-0903

Abstract

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.