Natural language generation of surgical procedures

A number of compositional Medical Concept Representation systems are being developed. Although these provide for a detailed conceptual representation of the underlying information, they have to be translated back to natural l anguage for used by end-users and applications. The GALEN programme has bee n developing one such representation and we report here on a tool developed to generate natural language phrases from the GALEN conceptual representat ions. This tool can be adapted to different source modelling schemes and to different destination languages or sublanguages of a domain. It is based o n a multilingual approach to natural language generation, realised through a clean separation of the domain model from the linguistic model and their link by well defined structures. Specific knowledge structures and operatio ns have been developed for bridging between the modelling 'style' of the co nceptual representation and natural language. Using the example of the sche me developed for modelling surgical operative procedures within the GALEN-I N-USE project, we show how the generator is adapted to such a scheme. The b asic characteristics of the surgical procedures scheme are presented togeth er with the basic principles of the generation tool. Using worked examples, we discuss the transformation operations which change the initial source r epresentation into a form which can more directly be translated to a given natural language. In particular, the linguistic knowledge which has to be i ntroduced-such as definitions of concepts and relationships-is described. W e explain the overall generator strategy and how particular transformation operations are triggered by language-dependent and conceptual parameters. R esults are shown for generated French phrases corresponding to surgical pro cedures from the urology domain. (C) 1999 Elsevier Science Ireland Ltd. All rights reserved.