Multi-Agent Cooperation Infrastructure to Support Patient-Oriented Telecare Services

Agent technologies are now being considered for automating tasks in e-commerce applications. However, conventional agent modules with predefined functions, but without the ability to modify behavior dynamically, may be too limited for mediating telecare medical services properly, since they cannot switch roles or adjust their behavior to support dynamically-formed healthcare systems. In this paper we propose a design and an implementation of an infrastructure that is able to collect and analyze large amount of heterogeneous data to support patient-oriented telecare services. We have constructed a Java-based dynamic infrastructure that utilizes XML-formatted data to enable dynamic behavior modification of agent. The key characteristic of our work is that it "follows the patient" encompassing three correlated stages: patient in ex-life, patient in time, patient in action. Although the agent module is under further development it has already produced clinically- acceptable results utilizing data from 86 subjects of the Cardiology Department of the University Hospital of Patras.