UNDERSTANDING CLINICAL NARRATIVE TEXT

A linguistic analysis of the narrative which appears in patient docume nts, in the domain of urology, has shown that the information is conve yed by a small number of semantic statement types composed of syntacti c combinations of medical and English word classes. It has been shown that words from different classes occuring in the diagnosis for the pa tient disease follow some regular grammatical rules. This impressive l inguistic regularity in clinical narrative has made it encouraging to develop a recognizer which is capable of recognizing disease diagnosis in the narrative clinical documents in the domain of urology. The rec ognizer was implemented using the C language. This recognizer output i nformation which could be refined to a form suitable for management by a database management system for further computerized information pro cessing.