On the application of multi-agent systems in Health Care

On the application of
multi-agent systems in
health care
Dr. Antonio Moreno
ITAKA-Intelligent Tech. for Advanced Knowledge Acquisition
Computer Science and Mathematics Dep.
Univ. Rovira i Virgili, Tarragona
http://deim.urv.cat/~itaka
Sevilla, June 14th-15th 2010

Outline of the talk

Introduction
Health care areas where agents have
been applied
K4Care: agent-based provision of Home
Care services
Benefits of the application of agents in
health care problems

Characteristics of Health Care problems
Distributed knowledge
E.g. different units of a hospital
Coordinated effort
E.g. receptionist, general and specialised
doctors, nurses, tests personnel, ...
Complex problems
E.g. patient scheduling
Need to manage different types of
knowledge
Medical, organisational, procedural

Why use MAS in Health Care?

MAS are inherently distributed
Agents can coordinate their activities, while
keeping their autonomy and local data
Dynamic and flexible distributed problem
solving mechanisms
Use of personalisation techniques
Knowledge-based behaviour of the system

Main application domains
5

1‐Medical data management 3‐Planning and resource allocation
National electronic Library for Health (NeLH) Agent.Hospital (simulation)
Management of palliative patients (PalliaSys) CARREL (transplant management)
Virtual Electronic Patient Record (VEPR) Medical Information Agents (MIA)
Context‐aware Hospital Information System (CHIS) Medical Services Coordination (MeSSyCo)

5‐Composite Systems
SHARE‐IT (assistive technologies)
K4Care (Home Care)
Geriatric Ambient Intelligence (GerAmI)

2‐Decision support systems 4‐Remote care, telemedicine
Singh’s intelligent assistant Aingeru (elder monitoring)
HealthAgents [Microart] INCA (Community services)
Health Care Services (HeCaSe) Medical Contact Centres [Koutkias]
Monitor chronic patients [Cervantes]

1-Medical data management
This area includes different kinds of systems:
Information agents that collect, filter and analyse
medical information available in electronic
resources
Agents that provide a transparent access to
physically distributed information sources
Different medical organisations, or different units within
a hospital
In general, intelligent management of the personal
and medical information in the Electronic Health
Record of a patient

2-Decision support systems
These systems aim to assist the professionals in
the decisions to be taken during the healthcare
process.
E.g. help in the diagnosis phase, or in the definition and
execution of the most appropriate personalised
treatment.
Agents can also perform routine actions, such as
checking periodically the patient state, or waiting
for the results of a medical test to be available
They usually perform complex reasoning
processes

3-Planning and resource allocation

Systems in which professionals and/or
medical resources are represented by
autonomous agents
The basic aspect is the coordination of their
activities to take appropriate decentralised
scheduling decisions in medical centres
E.g. patient scheduling

4-Remote care, Telemedicine

In most cases, the basic aim of the system is
to monitor continuously the state of the
patients, allowing permanent care
Main elements of the system:
Collection of sensors
Analysers of signals from sensors, problem
detection
Generation of alarms and reports for medical staff

5-Composite systems

Agent-based platforms that integrate
different ICTs and Artificial Intelligence
techniques in order to provide an efficient
coordination of the activities to be performed
to provide an efficient health care to a
particular kind of patients

K4Care European project

2006-2009, 13 partners, coordinated by
URV
The aim of the K4Care European
project was to provide a Home Care
model, as well as to develop a prototype
system, based on Web technology and
intelligent agents, that provided the
services defined in the model

K4Care Model: Structure
1 Nuclear Structure + n Accessory Services

THE K4CARE MODEL

...

HCNS

Actor Service
Data/Information
Action Procedure

K4Care Model: Actors and Teams

K4Care Knowledge structures

EHCR: Electronic Health Care Record
APO: Actor Profile Ontology
CPO: Case Profile Ontology
Procedures
FIP: Formal Intervention Plan
IIP: Individual Intervention Plan

K4Care Ontologies (I)

Actor Profile
Ontology (APO)
Types of actors
Actions related to
each role
Platform services
Procedures
Documents
...

K4Care Ontologies (II)
Case Profile Ontology (CPO)
Diseases
Syndromes
Signs and symptoms
Social issues
Assessment tests
Interventions
...

Procedures, FIPs and IIPs
All the careflow procedural aspects are
represented in SDA* (States, Decisions, Actions)
Procedures are formal specifications of the way in
which an administrative service (e.g. admit a new
patient to the Home Care service) has to be
implemented
Formal Intervention Plans (FIPs) are formal
structures representing the health care workflow to
assist patients suffering form particular ailments or
diseases
Problem: application of guidelines to co-morbid patients

Definition of an
Individual Intervention Plan

K4Care platform features
Agent-based Web-accessible platform that
provides a set of basic Home Care services
Admit a patient to the Home Care service
Create an Evaluation Unit
Assign an Evaluation Unit to a particular patient
Assess the initial state of a patient
Definition of an IIP for a patient
Apply IIP to the patient
...

Multi-agent system
1 Actor Agent for each user, permanently
running
When the user logs in, a Gateway Agent is
dynamically created
Two-way communication Web-servlet-GA-AA
When an Actor Agent has to manage the
execution of a procedure/IIP, it creates
dynamically a SDA-executor Agent

Agent-based execution of IIPs (I)

Agent-based execution of IIPs (II)

Agent-based execution of IIPs (III)

Agent-based execution of IIPs (IV)

Summary of K4Care main aspects
Declarative (medical, organizational) and procedural
knowledge
Web-based interaction between agents and end-
users
Individual Intervention Plans allow practitioners to
implement accurate and personalised sequences of
actions for the treatment of a particular patient
The architecture allows implementing agent-based
coordination methods between the actors relevant in
Home Care, which adapt their behaviour
dynamically depending on the knowledge available
in the platform

Positive aspects of MAS [in HC] (I)
Modularity
A complex problem is divided in subproblems
which may be solved by autonomous units, with
the appropriate coordination among them
Efficiency
Agents may be running in different computers,
speeding up the resolution of the problem
Decentralisation
Less single point-of-failure risk than centralised
systems

Positive aspects of MAS [in HC] (II)
Flexibility
The MAS components may change at run time,
the tasks may be dynamically distributed
Personalisation
Personal agents may have information on the
user preferences and adapt the system’s
behaviour to them
Distributed planning
Use of coordination techniques for distributed
problem solving

Positive aspects of MAS [in HC] (III)
Monitoring, alarm management
Continuous monitoring of personal and medical
data, with immediate activation of emergencies
when needed
Proactivity
Agents may perform tasks without requiring a
constant intervention or request from the user
Security
Confidentiality of medical data

Recommended extra material
Publications at http://deim.urv.cat/~itaka
D.Isern, D.Sánchez, A.Moreno
Agents applied in health care: a review
International Journal of Medical Informatics, Vol. 70, pp.
145-166, 2010

On the application of
multi-agent systems in
health care
Dr. Antonio Moreno
ITAKA-Intelligent Tech. for Advanced Knowledge Acquisition
Computer Science and Mathematics Dep.
Univ. Rovira i Virgili, Tarragona
http://deim.urv.cat/~itaka Sevilla, June 14th-15th 2010

Agents – general definition

Computational entities, capable of sensing
the environment and acting proactively and
autonomously upon it in order to satisfy their
design objectives
Can communicate with other agents to share
information, coordinate their activities and
cooperate to solve complex distributed
problems

DBs, Electronic Health Care Record

Data Base: with information about the
K4Care actors as users of the K4Care
Platform (e.g. contact information)
EHCR: with the data about the Home-
Care processes performed within the
K4Care Platform
Medical documents stored in XML

FIP for the
management
of
hypertension

On the application of multi-agent systems in Health Care

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to On the application of multi-agent systems in Health Care (20)

More from Antonio Moreno (19)

Recently uploaded (20)

On the application of multi-agent systems in Health Care