The major objective of the project is the development of methods for the automatic processing of natural medical language. This involves both development of algorithms for parsing and representing natural text, and creation of computer dictionaries and other data bases for medical specialties such as surgical pathology. Research in medical linguistics continued on the development of a theorem prover for a higher order predicate calculus, constrained by assumptions associated with relevance/entailment logic, as a vehicle for parsing in a head-feature based grammar formalism. Collaboration continued on the Clinical Information Utility with the Laboratory of Pathology, NCI/DCBD/LP, and with DCRT/DMB, in using LSM's automatic encoding system to maintain and improve the data base of Clinical Center surgical pathology reports. Transition from word processors to the Clinical Center MIS for producing the surgical pathology reports is near completion. Development of the Lexicographic Environment Software continued but at a reduced pace. This is designed to provide medical lexicographers and medical language researchers with a comprehensive tool for examining data and testing hypotheses, as well as for constructing both semantic and syntactic features of medical dictionaries.
