Collaborative Medical Software Agents

A Collaborative Software agents framework to support medical decision process based on ontology and bayesian networks is presented. The framework has three key components: knowledge management, uncertainty reasoning and software agents. The ontology is chosen to address the management of different medical sub-domains knowledge. The Bayesian Network is the graphical model for probabilistic knowledge representation and reasoning about partial beliefs under uncertainty. Each medical sub-domain will be embedded into a software agent allowing collaboration between them in terms of sharing information. The purpose of the framework is to allow the automated information exchange between different medicine specialists based on shared concepts within specific areas of expertise. Each sub-domain expert will act over his concepts (evidences, causes, effect) and will receive the information from other sub-domains as well. The agents will be in charge of belief propagation over the entire domain, among all concepts.