Agent Technology and Healthcare: Possibilities, Challenges and Examples of Application - Computational Intelligence in Healthcare: Advanced Methodologies - page 22

coordination of several healthcare entities and to provide personalised healthcare
services. Section 5 is devoted to K4Care, a knowledge-driven agent-based platform
which provides personalized home-care services for elderly patients following a spe-
cially designed European Home Care model. The chapter finishes with some general
conclusions on the use of agents in health care, focusing on some lines of future work
that could help to move agents from the academic labs to their routine use in clinical
settings.

2 Agents and Multi-agent Systems

In the last twenty years, agents have emerged as a promising computer engineering
paradigm. In one of the most relevant works in the area, Wooldridge [52] defined an
agent as follows:

“An agent is an entity that must be able to perceive the physical or virtual world
around it using sensors. A fundamental part of perception is the ability to recognize
and filter out the expected events and attend to the unexpected ones. Intelligent
agents use effectors to take actions either by sending messages to other agents or by
calling application programming interfaces or system services directly”

Agents represent a core idea in modern computing, where cooperation and communi-
cation are heavily stressed. The agent paradigm advances the modelling of techno-
logical systems, as agents embody a stronger and more natural notion of autonomy
and control than objects.
As software entities, agents implement their functionalities in the form of behav-
iours which are initiated by means of an external stimulus ( e.g., the reception of a
signal from a physical entity or a message from another agent) and execute a series of
actions according to the agent's objectives. Those objectives are designed to mimic
the behaviour of a real world entity ( e.g., a user, a physical resource) that the agent
represent, framed in the context of a complex and distributed system. In order to offer
a realistic implementation of a real behaviour and an added value to the final user,
agents typically incorporate Artificial Intelligence techniques ( e.g. , data mining, plan-
ning techniques, fuzzy logic, etc.) in order to decide the actions to execute in function
of the input data or to pre-process or to filter information according to the user profile
to which interact.
In order to extend the theoretical definition of agents, Wooldridge and Jennings
[53] distinguished the main characteristics and features that intelligent agents may
exhibit:

• Autonomy . Agents can operate without the direct intervention of humans or oth-

ers entities, and may have some kind of control over their actions and internal
state.

• Reactivity . Agents may perceive their environment (physical world, a user, a col-

lection of agents, the Internet, or a combination of all of them) and respond in a
timely fashion to changes that occur in it. Those changes may be transmitted to
the agent in the form of stimulus, signals or messages.