A framework for building cooperating agents

There is an increasing awareness in the AI community that the dissemination of Decision Support Systems in medical practice strictly depends on their seamless integration with those tools already available at clinical settings. In order to achieve that goal, this paper further elaborates on a previous research conducted by the same authors which identified the main components of a Distributed Hospital Information System based on Cooperating Software Agents. We propose a computational architecture supporting communication and the occurrence of multiple conversations among agents, thus effectively enforcing their cooperation. The architecture is being tested for implementing a prototype system able to coordinate the joint efforts of the several health care providers involved in managing patients affected by Acute Myeloid Leukemia.