Agent Technology and Healthcare: Possibilities, Challenges and Examples of Application - Computational Intelligence in Healthcare: Advanced Methodologies - page 21

sources, the security issues related to the management of sensitive data and the ro-
bustness in front of possible errors are just some of the difficulties which a computer-
ized healthcare system has to face.
Agent technology has emerged in the last 20 years as a promising paradigm for the
modelling, design and development of complex systems. Agents and, more generally,
multi-agent systems [47] allow to model in a realistic way complex, heterogeneous
and distributed systems and environments. Agents offer a natural solution in which
each of the human actors or physical entities are modelled by means of an agent. In
the first case, for example, agents may be responsible of storing the human's knowl-
edge, the actions he/she is allowed to perform, the interactions with other actors, and
the access and processing of data. In the second case, agents may control the access to
the resources, control timetables or implement planning tasks.
Internally, each agent implements the behaviours, the main communicative proc-
esses and the methods needed to mimic real-world processes. Moreover, agents can
take their own decisions, based on their internal state and the information that they
receive from the environment.
When several agents cooperate in the same environment, a multi-agent system may
use distributed problem solving techniques, allowing to decompose a complex prob-
lem in several pieces, which can be concurrently solved in a distributed and decentral-
ised fashion. As a result of the distributed cooperation of several heterogeneous
agents, global complex behaviours emerge, allowing realistically modelling and simu-
lating real-world environments, which may be difficult to simulate using other com-
putational paradigms.
As will be argued in section 3, all those possibilities make intelligent agents and
multi-agent systems an ideal paradigm to model or simulate healthcare processes.
Healthcare entities can be modelled as agents, implementing appropriate behaviours
and developing high-level negotiation and communication processes, which represent
faithfully real-world medical interactions. The use of intelligent agents in Medicine
may be considered a complementary technique to improve the performance of typi-
cally ad-hoc and closed medical systems, in terms of interoperability, scalability and
flexibility [36].
The contents of this chapter are based in the authors' experience on applying agent
technology to healthcare processes. It offers a comprehensive introduction to agent
technology from the point of view of a newcomer, illustrating the explanation with
several practical examples of agent-based healthcare systems in which authors have
been closely involved. As a result of this analysis, several challenges in which re-
search efforts should be put in the following years in order to facilitate the transition
of agent technology from the prototypical and research scope to a real-word setting
are presented.
The rest of the chapter is organised as follows. Section 2 makes an introduction to
the main properties and characteristics of agents and multi-agent systems from a theo-
retical point of view. Section 3 argues why agent technology is a good option to be
applied in the healthcare environment, and presents the main medical areas in which
agents have been applied in the last 10 years. Several relevant works applying agents
in those areas are also introduced. The benefits that agents can bring to the healthcare
domain are also illustrated with two practical examples. Section 4 describes HeCaSe2,
a multi-agent system which is able to enact clinical practical guidelines by the