Automated information extraction from free-text medical documents for stroke key performance indicators: a pilot study

Intern Med J. 2022 Feb;52(2):315-317. doi: 10.1111/imj.15678.

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.

Keywords: key performance indicator; machine learning; natural language processing; random forest.

MeSH terms

  • Electronic Health Records
  • Humans
  • Information Storage and Retrieval
  • Machine Learning*
  • Natural Language Processing
  • Pilot Projects
  • Stroke* / therapy