Healthcare Expert System based on the Group ... - CiteSeerX

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 {rmmateo, bgerardo, jwlee}@kunsan.ac.kr

Abstract Expert systems in healthcare services are widely studied by researchers where the accuracy on diagnosing and efficiency of the system are the considerations. Also, recent researches include decision support services, expert medical services and autonomous management 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 and provide an advance treatment and prevent loss of life. This research shows an object implementation of the healthcare expert system (HES) based on the group cooperation model. An object performs data mining for specific disease diagnosing tool used by the expert. The proposed agent managers coordinate the processing of client requests. Replicas of object in servers are introduced to provide quality of service for clients. An adaptive scheme managed by the load balancing service is presented. The implementation of the healthcare expert system using the group cooperation model is presented.

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 multiagents 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 developmentallydelayed 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 subcomponent 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.

object where it can act as an identifiable, encapsulated entity that may be invoked by a client.

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

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

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.

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. 5. Interaction of the load balancing service

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

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.

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. 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. 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. 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.

[5]

[6]

Fig. 12 Global load monitoring updating the loads from each objects

[7]

5. Conclusions and Future Work

[8]

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

[9]

[10]

[11]

[12]

ions in Medical Images” IEEE Computational Science a nd Engineering, Vol. 2, No. 5, 2000, pp. 39-45 Verma, B.; Zakos, J., “A Computer-aided Diagnosis System for Digital Mammograms based on FuzzyNeural and Feature Extraction Techniques” IEEE Transactions on Information Technology in Biomedicine, Vol. 5, No. 1, 2001, pp. 46-54 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 Schetinin, V., “Neural Network based Agent for Discov ering Rules in Medical Databases” available at http://cit eseer.ist.psu.edu/update/605607 Negnevitsky, M., Artificial Intelligence, A Guide to Intelligent Systems Second Edition, Addison Wesley pp. 25-54 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, 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 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 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.