The problem of linguistic approximation in clinical decision making

✍ Scribed by Rosanna Degani; Giovanni Bortolan

• Publisher: Elsevier Science
• Year: 1988
• Tongue: English
• Weight: 872 KB
• Volume: 2
• Category: Article
• ISSN: 0888-613X
• DOI: 10.1016/0888-613x(88)90105-3

✦ Synopsis

This paper deals with the problem of linguistic approximation in a computerized system the context of medical decision making. The general problem and a few application-oriented solutions have been treated in the literature. After a review of the main approaches (best fit, successive approximations, piecewise decomposition, preference set, fuzzy chopping) some of the unresolved problems are pointed out. The case of deciding upon various diagnostic abnormalities" suggested by the analysis of the electrocardiographic signal is then put forward. The linguistic approximation method used in this situation is finally described. Its main merit is its simple (i.e., easily understood) linguistic output, which uses labels whose meaning is rather well established among the users (i.e., the physicians).