Natural Language Generation in Health Care
- Affiliations of the authors: Department of Computing and Electrical Engineering, Heriot-Watt University, Edinburgh, Scotland (AJC); Computing and Information Science Department, University of Pennsylvania, Philadelphia, PA (BLW); Department of Public Health, University of Glasgow, Glasgow, Scotland (RBJ)
- Correspondence and reprints: Alison Cawsey, Department of Computing and Electrical Engineering, Heriot-Watt University, Edinburgh EH14 4AS, Scotland. E-mail: alison{at}cee.hw.ac.uk
- Received 29 May 1997
- Accepted 17 July 1997
Abstract
Good communication is vital in health care, both among health care professionals, and between health care professionals and their patients. And well-written documents, describing and/or explaining the information in structured databases may be easier to comprehend, more edifying, and even more convincing than the structured data, even when presented in tabular or graphic form. Documents may be automatically generated from structured data, using techniques from the field of natural language generation. These techniques are concerned with how the content, organization and language used in a document can be dynamically selected, depending on the audience and context. They have been used to generate health education materials, explanations and critiques in decision support systems, and medical reports and progress notes.
Footnotes
-
This work has been supported in part by the National Library of Medicine under grants R01 LM05764-01 and RO1 LM06325, and the Engineering and Physical Sciences Research Council under grant GR/K55271.
-
↵* A more detailed and up to date survey is available online at web address: http://www.cse.ogi.edu/CSLU/HLTsurvey/HLTsurvey.html, chapter 4.
-
↵† Mail merge systems, available with most word processors, allow information from a database to be incorporated into a text document in simple ways, allowing, for example, mass production of personalised letters using information from a customer database. Simple IF-THEN statements often allow different chunks of text to be output depending on the current data.
