Web Service to Support the Prostate Cancer Diagnosis using Expert System

Marcos Aurélio Pereira, Jefferson Luiz Brum Marques
Institute of Biomedical Engineering, Florianopolis, Brazil

Abstract
Web services, as the name implies, are services offered via the Web. In spite of the Web services and consumers of Web services are typically businesses, making Web services predominantly business-to-business (B-to-B) transactions, its use can also be applied for other purposes besides the businesses world, like services of information to support decision making procedures, search and change of information, and data processing in expert systems. The utilization of Web Service architecture for a system applied to support the decision in the medical area allows that several doctors can use the system in a simultaneous and independent way, making possible the concentration of clinical cases in a single database. However, integrity and security factors are extremely necessary to guarantee only permitted access to the patients' clinical data. The concentration of several clinical cases in a unique database and under constant updating, allow a contribution study about the behaviour of the disease in the most different cases, and the search of approaches to improve the efficiency of its treatment.

Keywords: Web Services, Prostate Cancer Diagnosis, Expert System

1 - Introduction

Web Services is a term that has come to be used in many contexts because there are many things that might be called "Web services" [14][6]. However, most people tend to associate with one definition: services that were used over the Web and was intended to encompass such things as applications distributed across the network, dynamically brought together at run-time [6].
Web services are the ultimate unification of the Web, databases, XML (a text-based markup language that is fast becoming the standard for data interchange on the Web [7]), and distributed systems. This is the way for bringing electronic data interchange, transaction systems, and business-to-business services into the 21st century [13]. The web services architecture is ideal for systems that request constant improvement in the algorithm of processing of the information, mainly in decision support systems through expert systems, because this processing is supplied as a service. Beyond this, the concentration of the data in a single database, together with a high security system and a workflow characteristic allows that client users of the system may run future services such as statistical evaluations for scientific studies and the use of techniques based on neural networks, without necessity of installation or configuration.
A lot of information about the prostate cancer comes from the study of several clinical cases. An example of that is the fact that prostate cancer is more common in North America and north western Europe. It is far less common in Asia [3].
In spite of the objective of the system is to give to the doctor decision support information through an expert system, that is, the expert system will inform with base in clinical data, if the patient tends to have a cancer or other type of prostate disease.

2 - Prostate Cancer

The prostate is a walnut-sized gland that forms part of the male reproductive system. It produces alkaline fluid that comprises approximately 70% of the seminal volume [1].

Prostate cancer is one of the most common cancers among males and the second most frequent cause of cancer deaths in men. With the development of prostate-specific antigen (PSA) screening, more men are identified earlier as having prostate cancer. While prostate cancer can be a slow-growing cancer, thousands of men die of the disease each year [2].
Prostate cancer is usually without symptoms in its early stage when it is curable. These early stage cancers can take up to ten years before the disease becomes life threatening without treatment. However, some prostate cancers can grow and spread quickly [3]. Prostate cancer when moves forward, it can be disseminated by the body, coming to reach other organs, and mainly, the bones. A pain in the spine in a patient in the risk age can be a dissemination of the tumour. It can also reach the ribs, basin, femurs, and other bones. Many times the patient has a spontaneous fracture of the femur, without any trauma, what can be a pathological fracture provoked by the dissemination of the tumour [4]. It can cause urinary symptoms when it moves forward until the urethra or vesical neck evolution, which is several times faster than BPH (Benign Prostate Hyperplasia), and without other previous urinary symptoms. Progression and obstruction of the ejaculatory ducts can result in haemospermia and decrease of the ejaculatory volume. The metastic involvement of the axial skeleton causes bony pain and anaemia for infiltration of the bone marrow. Approximately 50% of the diagnosed patients developed metastic disease, and consequently incurable, the prevention of this development with early diagnosis and treatment of the disease is a strategy that can save lives. Studies show that prostate cancer can be detected confined to the organ in 90,2-93%, and it is that objectives that the methods for the precocious diagnosis should be used [4].

3 - Web Services

A Web service is a software system designed to support interoperable machine-to-machine interaction over a network. It has an interface described in a machine-processable format. Other systems interact with the Web service in a manner prescribed by its description using SOAP-messages, typically conveyed using HTTP (HyperText Transport Protocol) with an XML (eXtensible Markup Language) serialization in conjunction with other Web-related standards. A Web service is viewed as an abstract notion that must be implemented by a concrete agent. The agent is the concrete entity that sends and receives messages, while the service is the abstract set of functionality that is provided [14].
In typical scenery of web services, the application sends a request for a service for a specific URL using the protocol SOAP (Simple Object Access Protocol) on HTTP. The service receives the request, it processes the request, and the answer comes back [7]. The protocol SOAP used in that process is a destined technology to utilization of distributed objects. This protocol has the differential of being simple and "Firewall-Friendly", in other words, it does not suffer security restrictions imposed by most of the firewalls in the Web, because it operates on standard protocols of Internet, like the HTTP and UDP. While its predecessors needed sophisticated executions, SOAP operates on the Web, and its messages (requisitions and answers) are simply XML documents [8].

4 - Expert System

Expert System consists of a system of artificial intelligence created to solve problems in a certain domain, where the whole knowledge is supplied by people that are experts in that domain, and the information where the expert system is based needs to be actual and accurate in this domain [10]. For this, is necessary a wide study of the problem domain to be solved, and how and which information the human expert use for solve it. With this information, a rule set is defined. These rules represent the knowledge of the human expert.
For the prostate cancer diagnosis, initially an urologist evaluates the patient. In case of suspicion, he is directed to the biopsy. This process can take until years until the confirmation of the disease, however once confirmed the cancer the patient is also accompanied and treated by the oncologist. Due to this, the system should supply support for the storage and processing of the clinical data in each one of these areas, synchronized and integrated in one system and database. Because this interaction, not just the knowledge of the urologist is important, but also of the oncologist and of the biopsy doctor.

5 - Web Service Using Expert System

The services provided by a service provider are accessed via a Web application, which is a dynamic extension of a Web server. This project uses Java™ 2 technology to do that. In the Java 2 Platform, Web components provide the dynamic extension capabilities for a Web server. Web components are supported by the services of a runtime platform called a Web container. The Web components are Java Servlets and JavaServer Pages (JSP). Servlets are Java programming language classes that dynamically process requests and construct responses. JSP pages are text-based documents that execute as servlets but allow a more natural approach to creating static content [7][5]. The logic of supplied service is implemented using Enterprise JavaBeans (EJB) technology, considered the middleware layer of the system. It is a server-side component encapsulates the business logic of an application. The execution of enterprise beans for the application is managed by EJB Container. Enterprise beans and their container run on the J2EE server [5].

With this, the service of expert system is a service supplied by the service provider, together with others services that are services of support to expert system, where this includes services of storage, processing, alteration and search for information. For this, it is necessary the utilization of a database, that store the data to be registered, and also give support of requested data by the application and use for the expert system. This project uses a mature and reliable database server with support to several requisitions and transactions at the same time.
The system is developed in multi-tier, in that each layer has a specific characteristic, and the communication between each one of them is accomplished through its interfaces. This division in layers allows better evolutionary maintenances, and each layer can be improved independently of others, but its interfaces must be respected. However, to guarantee quality and robustness of the system, designs were adopted in the conception and development of the project, allowing besides the quality factors a larger maintenance easiness, mainly the evolutionary maintenance, which will allow that improvements and new resources are more readily incorporated to the system.

The simple fact of the system to use Web services, it already turns necessary the use of the pattern Business Delegate. This pattern manages the complexity of distributed component lookup and exception handling, and may adapt the business component interface to a simpler interface for use by views [12]. The middleware layer uses Service Locator pattern and Session Facade pattern. The Service Locator pattern centralizes distributed service object lookups, provides a centralized point of control, and may act as a cache that eliminates redundant lookups. It also encapsulates any vendor-specific features of the lookup process. Already the Session Facade pattern defines a higher-level business component that contains and centralizes complex interactions between lower-level business components and decouples lower-level business components from one another, making designs more flexible and comprehensible [11][12] .
The Web application uses the MVC (Model-View-Controller) pattern that separates design concerns (data persistence and behaviour, presentation, and control), decreasing code duplication, centralizing control, and making the application more easily modifiable. The MVC architecture has its roots in the Smalltalk language, where it was originally applied to map the traditional input, processing, and output tasks to the graphical user interaction model [16]. However, it is straightforward to map these concepts into the domain of multi-tier applications [12][15]. The MVC architecture divides the application into three layers [15]:

" The Model represents the data and the business rules that govern access to and updates of this data. Often the model serves as a software approximation to a real-world process, so simple real-world modelling techniques apply when defining the model.

" The View renders the contents of a model. It accesses the data through the model and specifies how that data should be presented. It is the presentation layer and interaction with the user.

" The Controller defines application behavior. It translates interactions with the view into actions to be performed by the model. It dispatches user requests and selects views for presentation. In an Web application that uses Web services, this layer represents the bridge between the view layer through the models and the middleware layer that store and process the data in the model.

6 - Expert Service

To the expert system service to provide a result, it is necessary to insert previously the clinical data of the patient in the system through the web application, which supplies for the user (doctor) all the forms for data input. The clinical data of the patient can be acquired along the time, as well as a expert human, the expert service also needs a minimum amount of data about the patient for the system to calculate with base in those data, if the patient tends to have cancer or not.
This minimal data for evaluation of the patient includes the digital rectal exam (DRE) result that is evaluated by urologist, and the seric dosage of Prostatic Specific Antigen (PSA), that is evaluated in laboratory through a patient's blood sample [18]. As this data are minimal for evaluation, in cases of suspicion, the patient is indicated for the biopsy.
As more the clinical data of the patient are stored in the system, the expert system presents higher accuracy, to supply for the urologist based on those data, an approximate result on the tendency of prostate cancer.
The advantage of the system is that the several variables involved in the whole process of the patient's evaluation for the cancer confirmation or not, is stored and managed by the system, allowing to obtain easily values as PSA velocity and PSA density [17], which in the traditional process not always it is easy. With this, the storage of those clinical data of the patient, the automatic management of that information and the processing of the expert system allow a reduction of the time of evaluation for the confirmation of prostate cancer. This reduction of the time for the detection of the disease still in initial stage increases considerably the chances for patient's cure.

7 - Integrity and Security

The system makes possible the doctor to insert the patients' data together with dated diagnosis. Also, allows accompaniment of the patients' state through the exams and diagnoses, inclusive in cases of death due to the cancer, and the prognostic prediction through the expert system, where this option, based on the patient's data and processing based on rules, will help the doctor in his/her final evaluation.
However, due to the access through Internet, factors of security are very important because the data that are being carried by the net and processed are extremely confidential. Therefore, the system possesses a high integrity and security factor of the clinical data through protocols of authentication and use of the system. This resource allows only registered users of the system to have access to login, and users of the system will have only access to the data they are responsible. For example, the doctor will have access for visualization, accompaniment and updating of the data of his/her patients, and the visualization, accompaniment and alteration of patients' data of the others doctors of the system are inaccessible to him/her. A user with secretariat role has access only to the patients of the clinic where he/she works; the clinical data will be inaccessible to this user.

8 - Discussion

This system due its characteristics, allows the reduction of the time of the patients' diagnosis, increasing consequently their chances of cure; reduction of the costs for the patient due the possibility of smaller time for diagnosis; and it makes possible the doctor to treat and to accompany more patient.

9- Conclusions

Although the system is in development phase, this is a great tool of aid to the doctors, due to possibility of controlling all the variables involved in the process of the patient's accompaniment before his prognostic and to provide him through those factors and of the expert system the tendency of patients to have prostate disease. The distributed characteristic together with the centralization of the data of several clinical cases, allow a great scientific study of the disease in a multi centre way, and constantly updating the system database.

10 - References

[1] - LEVEILLEE, Raymond J. Benign Prostate Hyperplasia. eMedicine. Syracuse, Feb. 2003. Available at: <http://www.EMedicine.Syracuse.com/med/UROLOGY.htm>.
[2] - THEODORESCU, Dan. Prostate Cancer: Biology, Diagnosis, Pathology, Staging, And Natural History. eMedicine. Syracuse, Dec. 2001. Available at: <http://www.EMedicine.Syracuse.com/med/UROLOGY.htm>.
[3] - THE PROSTATE CANCER RESEARCH FOUNDATION OF CANADA. Ending the Threat of Prostate Cancer: Progress Report on Prostate Cancer Research. Toronto, Nov. 2002. Available at: <http://www.prostatecancer.ca>.
[4] - TAVARES, Yan Schaefer. Comparação dos achados da ultrassonografia transretal e resultados de biópsia na avaliação diagnóstica do câncer de próstata. Florianópolis: UFSC. 2002. Cap. 1, p. 1-5.
[5] - Sun Microsystems, Inc. The J2EE™ Tutorial. Sun Microsystems, Inc. Palo Alto, abr. 2002. Available at: <http://www.javasoft.com>
[6] - WATERHOUSE, Mark. Web Services Architect Review. Web Services Architect. [S.l]. may. 2002. Available at: <http://www.webservicesarchitect.com/content/articles/mark01print.asp>
[7] - Sun Microsystems, Inc. The Java™ Web Services Tutorial. Sun Microsystems, Inc. Palo Alto, mar. 2002. Available at: <http://www.javasoft.com>.
[8] - BRAZ, Márcio R. Tecnologia de Web Services: Definições e Perspectivas. Developer's Magazine, Rio de janeiro, ano 7. n. 80. p.22-23. abr. 2003
[9] - CHAMPION, Michael. FERRIS, Chris. NEWCOMER, Eric. ORCHARD, David. Web Services Architecture: W3C Working Draft 14 November 2002. W3C. Cambridge, nov. 2002. Available at: <http://www.w3.org/TR/2002/WD-ws-arch-20021114/>.
[10] - LEVINE, Robert I. DRANG, Diane E. EDELSON, Barry. Inteligência Artificial e Sistemas Especialistas: Aplicações e Exemplos Práticos. Tradução: Maria C. S. R. Ratto. Revisão técnica: José C. Damski. São Paulo: McGraw-Hill 1986.
[11] - Sun Microsystems, Inc. Core J2EE Pattern Catalog. Sun Microsystems, Inc. Palo Alto, mar. 2002. Available at: <http://java.sun.com/blueprints/corej2eepatterns/Patterns/index.html>.
[12] - Sun Microsystems, Inc. J2EE Patterns Catalog. Sun Microsystems, Inc. Palo Alto, mar. 2002. Available at: <http://java.sun.com/blueprints/patterns/MVC.html>.
[13] - LEA, Doug, VINOSKI, Steve. Middleware for Web Services. Institute of Electrical and Electronics Engineers, Inc. New York, 2003. DS Online ISSN: 1541-4922
[14] - BOOTH, David, HAAS, Hugo, MCCABE, Francis, NEWCOMER, Eric, CHAMPION, Michael, FERRIS, Chris, ORCHARD, David. Web Services Architecture: W3C Working Draft 8 August 2003. W3C. Cambridge, aug. 2003. Available at: <http://www.w3.org/TR/2002/WD-ws-arch-20021114/>
[15] - SINGH, Inderjeet, STEARNS, Beth, JOHNSON, Johnson. Designing Enterprise Applications with the J2EE™ Platform. 2nd Edition. New York: Addisob-Wesley, 2002. 416p.
[16] - BURBECK, Steve. Applications Programming in Smalltalk-80™: How to use Model-View-Controller (MVC). [S.l]. 1992. Available at: <ftp://st.cs.uiuc.edu/pub/Smalltalk/>
[17] - BROSMAN, Stanley A. Prostate-Specific Antigen. eMedicine. Syracuse, Jun. 2002. Disponível em: <http://www.EMedicine.Syracuse.com/med/UROLOGY.htm>
[18] - ASTRAZENECA PHARMACEUTICALS LP. Prostate Cancer: What it is and How it is treated. A Booklet for Patients. Wilmington. 2002. Available at: <http://www.prostatainfo.com>