Realising the vision of pervasive health care will generate new challenges for knowledge management and data integration. Such challenges are fundamentally different from issues and problems that we face in centralised approaches as well as non-clinical scenarios. In this paper, we reflect upon our experiences in the
project wherein a prototype system was developed to support data integration and decision making in the breast cancer domain. While the decision making needs to rely on different clinical expertise, the
system leveraged a system ontology to glue together distributed services. Situating the
system in a highly pervasive environment reveals the inefficiency of global vocabularies via domain ontologies and the inappropriateness of “static” system ontologies with assigned system configuration instances. We examine the capability of a process calculus based language, Lightweight Coordination Calculus (LCC), in meeting knowledge management challenges in pervasive health care. The key difference in approach lies in making the representational abstraction reflect the relative autonomy of the various clinical specialisms (
, mammography or histopathology) involved in contributing to patient management. The bringing together of diverse forms of information necessary for the collective medical assessment is managed by tracking the message passing protocols undertaken by medical personnel. The scope within LCC of accommodating boolean-valued constraints allows for flexible integration of heterogeneous sources in multiple formats, which are characteristic features of a pervasive healthcare environment.