Healthcare Expert System Based on the Group Cooperation Model

John

Abstract
Expert systems for healthcare services are widely studied where accuracy of diagnosis and efficiency of the system for various services are examined. Also, recent researches include decision support services, expert medical services and autonomous management, which are based on multi-agent systems. The cooperation of the agents is crucial in analyzing, and managing the data of patient to detect abnormal patterns in order to provide an advance treatment. This research shows that the healthcare expert system (HES) is implemented on the group cooperation model. Communications through the components are designed based on multi-agent. The proposed agent managers effectively coordinate the processing of client requests. Replica managers provide quality of service for clients by replicating objects. An adaptive scheme using round robin and fuzzy least load algorithms is managed by load balancing service.


1. Introduction

Applying  information  technology  in  healthcare  is
necessary  in  providing  fast  and  efficient  services  to
clients  and  effectiveness  of  the  healthcare  system.  
Healthcare  services  provide  information  of  the
diagnosis  and  continuous  monitoring  of  patient  to
acquire  immediate  response  and  save  lives  in  case  of
critical  conditions.  Expert  systems  in  healthcare  are
important  on  giving  correct  information  for  diagnosis
and  providing  immediate  medical  services.  Current
research in decision  support system  (DSS)  uses  multi-
agents  for  the  expert  systems  [1].  Algorithms  that  are
accurate  in  classifying  diseases  are  used  for  medical
diagnosis. Neural network is a common used technique
[2,3]. Successful application examples show that neural
diagnostic  systems  are  better  than  human  diagnostic
capabilities.  Moreover,  neural  network  are  used  to
analyze medical images [4,5]. Various approaches and
techniques  are  used  to  improved  diagnosis  in  medical
images,  including  mammography,  ultrasound  and
magnetic  resonance  imaging.  An  ontology-based
intelligent  healthcare  agent  is  used  for  the  respiratory
waveform  recognition  to  assist  the  medical  staff  in
judging  the  meaning  of  the  graph  reading  from
ventilators  [6].  Neural  network-based  agents  are  used
for discovering rules in medical  database [7].  Medical
databases, which consist of patient histories, specialist's
conclusions,  laboratory  results,  etc.,  are  typically
distributed  set  of  semi-structured  data,  and  because
agent technology is well-suited approach to develop the
medicine  decision  support  systems,  the  integration  of
neural network in the agent is necessary. However, the
classification  of  diseases  depends  on  tools  and
techniques that doctors and specialists currently use to
diagnose and treat the patients. Moreover, there are lots
of  data  mining  algorithms  that  are  used  specific  to
diagnose  effectively  a  patient.  These  distributed
algorithms  and  tools  are  needed  to  interoperate  and
coordinate,  also  give  the  expert  a  way  to  choose  the
appropriate algorithm on providing the diagnosis.
This  research  shows  the  object  implementation  of
the healthcare expert system (HES) based on the group
cooperation  model.  The  proposed  system  consists  of
object groups where an expert chooses the objects that
are appropriate for processing the request. A service is
consisted of one or more objects where a single object
is  a  specific  data  mining  tool.  An  agent  manager  is
proposed to coordinate the processing of client requests
and communicate with other agent for the efficient load
distribution. Replica manager handles the creation and
deletion  of  object  replicas.  Object  replicas  in  several
servers are introduced to provide quality of service for
clients. An adaptive load  distribution  scheme which is
managed  by  the  load  balancing  service  is  presented.
The  implementation  of  the  HES  using  CORBA  based
on the group cooperation model is presented.

2. Expert Systems in Healthcare

Expert  systems  are  computer  programs  that  are
derived  from  a  branch  of  computer  science  research
called  artificial  intelligence  (AI)  [8].  The  expert
systems  are  reserved  for  programs  whose  knowledge
base contains the knowledge used by human expert. AI
is  concerned  with  the  concepts  and  methods  of
symbolic references, or reasoning, by a computer,  and
how the knowledge used to make those inferences will
be represented inside the  machine. The area of human
intellectual endeavor to be captured in an expert system
is called task domain. Domain refers to the area within
which the task is being performed. Task refers to some
goal oriented, problem-solving activity.  
In the domain of healthcare, it is important that  the
system is accurate in diagnosing because it deals with a
life  of  a  person  where  a  slight  error  of  treatments  or
diagnosis  can  cause  death  which  cannot  be  changed.
There  are  various  techniques  on  implementing  the
expert system and almost uses accurate and established
algorithms.  Most  of  these  methods  use  data  mining
techniques. Data  mining approach  are  commonly  used
to  extract  potential  knowledge  from  large  amounts  of
data  in  the  processing  knowledge  discovery.  The
design of an expert  system for  the  diagnosis of  health
problem is presented [9]. The expert system knowledge
base is composed by a set of production rules written in
classic  bi-valued  logic  and  by  a  set  of  potential  facts.
Another research  paper  suggests a  model of  a  chronic
diseases  prognosis  and  diagnosis  system  integrating
data  mining (DM)  and  case-based  reasoning (CBR)  is
proposed  [10].  The  extraction  of  hidden  knowledge
among  medical  history  data  of  developmentally-
delayed  children  is  explored  [11].  A  decision  tree  is
constructed  to  classify  delay  levels  of  each  type
according  to  physical  illness,  and  association  rule  is
applied  to  locate  correlations  between  cognitive,
language,  motor,  and  social  emotional  developmental
delays.  The  design  of  the  proposed  healthcare  expert
system  (HES)  is  an  expert  system  for  healthcare
services where the  service is customized by objects to
choose the appropriate algorithm for the request.  

3. Group Cooperation Model of HES

Cooperation  is  the  behavior  of  the  system
components  working  together  for  attaining  the  same
goal.  The  group  cooperation  model  of  HES  uses  the
adaptive load distribution from the previous work [12].
The  architecture  of  the  proposed  healthcare  expert
system  (HES)  shown  in  Figure  1  is  designed  for  the
group cooperation model.


Fig.  1.  Group  Cooperation  Model  for  Healthcare  Expert
System
The global view of the  group  cooperation model is
used  for  the  HES  shown in  Figure  1  which  consist  of
three level parts. It also shows the intercommunication
of  the  clients,  object  groups,  load  balancing  service,
and  server  replicas.  On  the  top-level,  clients  are
represented as heterogeneous hardware like desktop PC,
notebook  and  PDA.  Clients  are  interconnected  by  a
network  and  software  components  communicate  to
agent  managers  using  the  Object  Request  Broker
(ORB).  Object  groups  are  shown  in  the  middle-level
part where it is managed by agent managers and has the
object references of each  object in the  server replicas.
Objects  included  in  the  object  group  are  used  to
process  the  client  request.  In  this  research,  object
groups  are  defined  services.  These  services  have  sub-
component which is  needed for  processing its service.
For example, a cardiologist service needs an object for
association  rule  mining  algorithm  to  associate  the
client’s input on the previous history of other patients.
Objects  reside  in  the  server  replica  shown  in  the
bottom-level  of  Figure  1.  Communication  is  managed
by the agent managers or  AMs.  The bottom-level part
consists of  server  replicas  and  load  balancing  service.
These  components  are  used  to  implement  the  load
distribution  and  dynamic  replication.  Servers  are
managed  by  replica  managers  (RM).  A  server  replica
refers to a single server contains variety of objects. The
load  balancing  service  is  consisted  of  global  load
monitor and  analyzer  which manage the  efficient  load
distributions to object groups.   

Fig. 2. Client consultation process in a single service of HES
Figure  2  shows  the  interaction  of  client  to  the
service. Agent managers (AMs) receive the request and
process  the  information  through  the  sub-component
objects.  After  processing,  a  return  output  of
consultation is send to the client.

3.1. Components of Group Cooperation Model  

3.1.1 Agent manager (AM) – the main component of
the  service.  This  does  all  the  management  procedure.
All the components are connected to the AM and other
AM from outside the group are coordinating.  

3.1.2. Security component – checks the validity of the
client. The security verifies if the client has the access
privilege to the object. The proposed system processes
this with the  encryption  module  of  the  security.  If  the
client is not allowed to access the object then it sends a
message that it requires a valid user id and password or
the request is not allowed at all. After the procedure, it
sends message to AM the result of verification.

3.1.3.  Healthcare  expert  service  –  composite  of  the
task  which  consists  of  a  several  object  services  to
provide the service to client. In this research, a service
is  also  considered  as  an  object  group.  The  service
consists of AM, security and object references.  

3.1.4. Replica manager (RM) – does the creation and
management of objects  in  the  server.  The  RMs  within
the  server  replicas  is  coordinating  to  other  RM  to
manage the replicated objects. If object that changed it
values  then  RM  of  that  object  communicates  through
other RM to change the value of the same object

3.1.5. Objects – sub-components of the services. These
sub-components  are  requested  by  the  AM  to  process
the  request.  Moreover,  the  object  references  are
managed  by  the  AM  to  be  grouped.  The  purpose  of
grouping  the  objects  is  to  gather  the  related  object
chosen by an expert. In the group cooperation model of
HES,  each  group  uses a  plurality of request.  This is  a
property of an object  group  to  behave  like a singleton
object where it can act as an identifiable, encapsulated
entity that may be invoked by a client.

3.1.6. Global load monitor – monitors the loads. Each
object replica is  registered  to the global  load  monitor.
Upon  receiving  the  loads,  the  global  load  monitor
checks the value of the current distribution to compare
on the threshold value. If the load distribution is greater
than the threshold, then it communicates to the AMs to
switch the scheme from round robin to fuzzy least load
algorithm [12]. The same procedure of switching from
fuzzy  least  load  to  round-robin  is  done  if  the  global
load  monitor  determines  that  the  load  distribution  is
lower than the threshold.

3.1.7. Analyzer – analyzes the load of each servers and
objects. This is a sub-component of the load balancing
service  which  intercepts  every  request  to  process  the
fuzzy least load algorithm for distribution of request to
the  replica  objects  and  determine  additional  object
replicas needed. By processing the values to the  fuzzy
system,  it  determines  the  appropriate  least  loaded
object  to  forward  the  request.  Finally,  it  sends  the
object reference and issue a location forward.

3.2 Interactions of the Components

AM  is  the  main  receiver  for  client’s  request.  The
load  balancing  service  consists  of  interactive
components  for  intercepting  and  forwarding  request.
The  default  scheme  of  the  load  distribution  is  round
robin. Each AM are aware of the current scheme it uses
by  sending  message  of  their  status.  Once  the  AM
receives  the  request,  it  sends  messages  to  other  AM.
The  content  of  the  message  is  location  of  the  objects
and server replica which it invokes currently. If another
AM  receives  a  request  then  it  begins  invoking  the
object that is not used by other object groups. Figure 3
shows the communication between AMs.

   
Fig. 3. Agent managers informs other agents of the objects it
currently accessing




Upon the request,  the  object  sends an update  loads
to  the  global  load  monitor.  An  increment  of  loads  is
done  from the global  load  monitor  and  after  servicing
the client, it sends again an update load to decrease the
load.  While  the  global  load  monitor  is  on  the  update
procedure,  it  compares  the  status  of  the  load
distribution  from  the  objects  if  it  already  exceeds  the
threshold  value  [12].  Figure  4  shows  the  updating  of
loads  from  the  objects  of  the  server  replicas  to  the
global load monitor.


Fig.  4.  Updating  the  loads  of  each  replica  to  global  load
monitor
After  invoking  the  object  by  the  AM,  the  object
sends a load update to global load monitor. If the value
of  load  distribution  is  greater  than  the  threshold  then
the  round  robin  scheme  is  switched  to  the  fuzzy  least
load  [12].  At  this  point,  request  of  AM  is  directed  to
the analyzer where it determines the suitable  object  to
and  procedure  is  done  in  the  fuzzy  system.  Analyzer
issues  LOCATION_FORWARD()  reply  to  the  AM
request  with  the  reference  object.  Client  sends  the
request  to  the  new  reference.  Figure  5  shows  the
interaction of the load balancing components.


Fig. 5. Interaction of the load balancing service
The  global  load  monitor  communicates  to
analyzer  that  it  needs  to  change  the  scheme  and  after
sending the information, the analyzer informs the AMs
to  change  the  scheme from round  robin  to  fuzzy  least
load. Also, the same method of informing the analyzer
of  changing  the  scheme  is  done  when  the  load
distribution  is  already  constant.  The  interaction  of  the
components  is  presented  in  Event  Trace  Diagram
(ETD) for more detailed view in Figure 6 and 7.


Fig.  6.  Round  robin  accessing  the  objects  from  replica
servers where
σ
  <
Φ

Figure 6  presents  the  procedure  of  the  round  robin
algorithm.  Round  robin  is  the  default  scheme  of  the
load distribution. The
σ
determines how well the loads
are  distributed  to  the  objects  and
Φ
  is  the  threshold
value  for  the  load  distribution.  In  Figure  6  case,  the
system status  implies  that  there  is  no  need  to  forward
the  request  because  the  system  resources  are  utilized
enough  in  distributing  the  loads.  The  switching  of
algorithm occurs only if the load distribution is greater
than the threshold which means that the variance of the
load  distribution  is  large.  Figure  7  shows  the  case  of
where  the  system  switches  to  fuzzy  least  load
algorithm.


Fig.  7.  Fuzzy  least  load  accessing  the  objects  from  replica
servers
σ
  >
Φ

Figure 7 presents the ETD of switching the scheme
to  fuzzy  least  load.  Agent  managers  are  redirected  to
the  analyzer  for  processing  the  least  loaded  server.
Server  loads  refers  to  the  number  of  current  requests
accessing  the  objects  in  the  server  while  the  object
loads refers to the current request accessing the object.
These  loads  are  analyzed  in  the  fuzzy  system.  After
analyzing the most candidate object, the analyzer sends



location forward to the AM with the object reference of
the  most  candidate.  In  fuzzy  least  load  scheme,  the
same procedure of updating the global load monitor is
done. The analyzer continues to compare the threshold
to  the  load  distribution  and  when  the  threshold  is
greater  the  load  distribution  then  the  system  scheme
switches again to round robin.

4. Implementation of HES

The experiment used the  Borland Visibroker  7.0  to
implement  the  proposed  system  in  group  cooperation
model  which  is  CORBA  compliance.  All  components
of the healthcare expert system were developed in Java.
The  proposed  round  robin  and  fuzzy  least  load
algorithms  is  encoded  in  the  agent  manager  and
analyzer, respectively. Figure 8 shows the initialization
of the agent manager where the portable interceptor  is
created to intercept the request of clients and Figure  9
shows  that  the  replica  manager  creation  of  replica
objects.


Fig. 8. Agent manager and its interceptor are initialized
                           
Fig.  9.  Replica  manager  initialization  and  creation  of  the
object replicas
After  initializing  the  AM,  the  load  distribution
scheme  is  operational.  A  client  requesting  for  the
consultation is needed to login. The verification of the
client’s  authentication  is  done  by  the  security
component.  In  Figure  10,  an  input  form  appears  after
the verification procedure is done. If the client is not a
valid user then it cannot proceed to open the service.


Fig. 10. An applet of the input form for the HES consultation
After  the  verification,  the  client  is  directed  to  a
service  of  the  HES  shown  in  Figure  10.  The  user  is
required  to  enter  his  or  her  profile,  health  conditions
and  other  inputs  to  continue  the  consultation.  The
inputs are  processed  on  the  service  of  the  HES  which
uses the objects specified by the expert. In Figure 11, a
result of consultation is shown where the HES provide
a result that the client has a 5.9% chances of having a
heart disease based on the inputs. Figure 12 shows the
result of monitoring the object loads by the global load
monitor.


Fig. 11. Result of consultation from the HES
Figure  12  shows  the  execution  of  the  global  load
monitor. Each replicated object from the server replicas
is  registered  to  monitor  each  load.  If  the  objects  are
requested by AM then it updates the load.





Fig. 12 Global load monitoring updating the loads from each
objects
5. Conclusions and Future Work

This  paper  presents  a  healthcare  expert  system
(HES)  based  group  cooperation  model  to  implement
QoS  and  efficiency  of  the  system  by  using  object
replicas and effective coordination of the system using
the  proposed  agent  manager.  The  group  cooperation
model was discussed for the proposed healthcare expert
system.  A  group  of  objects  or  service  is  managed  by
the  proposed  agent  manager  where  it  chooses  the
necessary  objects  from  the  replica  server.  The  global
load  monitor  and  analyzer  manage  the  flow  of  the
adaptive scheme to implement the efficient distribution
of loads. The replica manager handles the replication of
the objects. The adaptive load  distribution  is based  on
two  algorithms  which  are  the  round  robin  and  fuzzy
least  load  algorithm.  This  research  implements  the
group  cooperation  model  for  the  HES  implementing
the  efficient service  by  using  components  and  objects
interactions and provides QoS by using replication and
adaptive load distribution schemes.
The  future  study  is  choosing  the  appropriate  data
mining  tools  in  processing  the  consultation  which
needs more details on specific diseases.


References

[1]  Foster  D.,  et  al.  “A  Survey  of  Agent-Based  Intelligent
Decision  Support  Systems  to  Support  Clinical
Management  and  Research”,  1st  Intl.  Workshop  on
Multi-Agent  Systems  for  Medicine,  Computational
Biology,  and  Bioinformatics,  Utrecht,  Netherlands,
2005
[2]  Brause, R., “Medical Analysis and Diagnosis by Neural
Networks”, Medical Data Analysis, Springer-Verlag, 20
01, pp. 1-13
[3]  Joo, S., Moon, W. K., and Kim, H. C., “Computer-Aide
d Diagnosis of Solid Breast Nodules on Ultrasound with
Digital Image Processing and  Artificial Neural  Networ
k” Engineering in  Medicine  and  Biology  Society,  Vol.
1, 2004, pp. 1397-1400
[4]  Giger, M. L., “Computer-Aided Diagnosis of Breast Les
ions in Medical Images” IEEE Computational Science a
nd Engineering, Vol. 2, No. 5, 2000, pp. 39-45
[5]  Verma,  B.;  Zakos,  J.,  “A  Computer-aided  Diagnosis
System  for  Digital  Mammograms  based  on  Fuzzy-
Neural  and  Feature  Extraction  Techniques”  IEEE
Transactions  on  Information  Technology  in
Biomedicine, Vol. 5,  No. 1, 2001, pp. 46-54  
[6]  Lee,  C.  S.,  Wang,  M.  H.:  Ontology-based  Intelligent
Healthcare  Agent  and  its  Application  to  Respiratory
Waveform  Recognition”,  Expert  Systems  with
Applications,  Vol. 33, No. 3, Oct. 2007, pp. 606-619
[7]  Schetinin, V., “Neural Network based Agent for Discov
ering Rules in Medical Databases” available at http://cit
eseer.ist.psu.edu/update/605607
[8]  Negnevitsky,  M.,  Artificial  Intelligence,  A  Guide  to
Intelligent Systems Second Edition, Addison Wesley pp.
25-54
[9]  Jimeneza,  M.  L.,    Santamaríab,  J.  M.,  Barchinoa,  R.  ,
Laitac,  L.,  Laitad,  L.  M.,  Gonzáleza,  L.  A.  Asenj,  A.,
“Knowledge  Representation  for  Diagnosis  of  Care
Problems  through  an  Expert  System:  Model  of  the
Auto-care  Deficit  Situations”,  Expert  Systems  with
Applications, Article in Press,
[10] Huang, M. J., Chen, M. Y., Lee, S. C., “Integrating Data
Mining  with  Case-Based  Reasoning  for  Chronic
Diseases Prognosis and Diagnosis”, Expert Systems with
Applications, Vol. 32, No. 3, April 2007, pp. 856-867
[11] Chang,  C.  L.,  “A  Study  of  Applying  Data  Mining  to
Early  Intervention  for  Developmentally-Delayed
Children”,  Expert  Systems  with  Applications,  Vol.  33,
No. 2, Aug. 2007, pp. 407-412
[12] R.  M.  A.  Mateo,  I.  Yoon  and  J.  Lee,  “Cooperation
Model  for  Object  Group  using  Load  Balancing”,
International  Journal  of  Computer  Science  and
Network Security, Vol. 6, No. 12, 2006, pp. 138-147.



Healthcare Expert System based on the Group Cooperation Model


Romeo Mark A. Mateo, Bobby D. Gerardo and Jaewan Lee
School of Electronic and Information Engineering, Kunsan National University
68 Miryong-dong, Kunsan, Chonbuk 573-701, South Korea