1 / 33

INTELLIGENT WEB-BASED EXPERT SYSTEM FOR EARLY DIAGNOSIS OF CANCER

Agenda. The ProblemThe SolutionResearch ObjectiveResearch ScopeDomain ExpertRelated WorkMethodologyWEBCANDI: A Web-based Expert System for Cancer DiagnoseConclusion. The Problem. Cancer is the fifth theme in the Phase 1 of the Ministry of Health's six years Healthy Lifestyle Campaign with th

yovela
Download Presentation

INTELLIGENT WEB-BASED EXPERT SYSTEM FOR EARLY DIAGNOSIS OF CANCER

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


    1. INTELLIGENT WEB-BASED EXPERT SYSTEM FOR EARLY DIAGNOSIS OF CANCER By Azizi Zakaria Fadzilah Siraj Nur Azzah Abu Bakar

    2. Agenda The Problem The Solution Research Objective Research Scope Domain Expert Related Work Methodology WEBCANDI: A Web-based Expert System for Cancer Diagnose Conclusion

    3. The Problem Cancer is the fifth theme in the Phase 1 of the Ministry of Health's six years Healthy Lifestyle Campaign with the slogan "Stay Ahead of Cancer". The theme was chosen as cancer is a rising problem in developing as well as developed countries. It is the second leading cause of death in most developed countries. Cancer has been reported as the fourth leading cause of death in Malaysia.

    4. The Problem Each year cancer affects at least nine million people world-wide and kills five million. The incidence of cancer in Malaysia is estimated to be around 150 per 100,000 population.

    5. The Problem This means there are about 35,000 new cancer cases each year. In the year 2000, there were 40,244 admissions for cancer in government hospitals.

    6. The Problem It is known that over one-third of cancers are preventable and another one-third is potentially curable if they are diagnosed early. Early detection improves the chances of curing cancer.

    7. The Problem Unfortunately, delays in detection of tumors are common among Malaysian cancer patients. A pilot study on cancer registration demonstrated that patients with stage 1 and 2 of the disease comprised 16.3% of all patients.

    8. The Problem As the number of new cases of cancer growing every year, Malaysia is facing a shortage of expert oncologist. Currently, there are only 28 oncologist in Malaysia with 7 of them working in government hospitals.

    9. The Solution We proposed an Expert System (ES) as the solution ES is a branch of Artificial Intelligence concerned with the design and implementation of intelligent program that are capable of emulating human cognitive skills such as problem solving, visual perception and diagnosis.

    10. The Solution Medical Expert System is primarily concerned with the construction of AI programs that perform diagnosis and make therapy recommendations.

    11. The Solution With the growing influence of World Wide Web (WWW), the expert system can be deployed on the WWW where the system is available anytime and easily accessed globally by large number of people. Thus, the web-based medical expert system can create the awareness of the primary prevention and early detection of cancer.

    12. Research Objective To utilize the power of Internet and web programming tools to spread the expert knowledge of a few highly skilled oncologist to the public To provide early diagnoses suggestions based on patients' symptoms available from any point on the Earth 24 hours per day 7 days per week.

    13. Research Scope

    14. Domain Expert Dr. Zabedah Othman Consultant Clinical Oncologist Department of Radiology and Oncology Kuala Lumpur General Hospital

    15. Related Work Cancer, Me?? Expert system for automated delivery of personal advice on how to reduce risk of cancer. It provides users with personalized cancer prevention information.

    16. Related Work ONCO-HELP is an expert systems for diagnosis of the general cancer diseases (Allahverdi et al.,2002). ONCO-HELP is a multimedia knowledge based decision support system for individual tumor entities. It makes individual and prognosis-oriented treatment of patient’s tumor possible

    17. Pereira et al. (2004), presents the development of a remote expert system in urological area to support the prostate cancer diagnosis. The prostate cancer is one of the most common cancers among men and the second most frequent death cause by cancer in men. Related Work

    18. Related Work Saritas et al. (2003), developed fuzzy expert system (FES) for diagnosing, analyzing and learning purpose of the prostate cancer diseases. FES used prostate specific antigen (PSA), age and prostate volume (PV) as input parameters and prostate cancer risk (PCR) as output. FES will determine the need for the biopsy and provide a range of the risk of the cancer diseases.

    19. The following are the activities involved in developing the Expert System: Knowledge Acquisition Knowledge Representation Design and Implementation Testing and Verification Methodology

    20. Components of WEBCANDI WEBCANDI divided into client component and server component. The client component can be accessed by using a web browser, we recommend Microsoft Internet Explorer, to ensure the run system run smoothly and correctly. The system can be access from the following URL, http://azportal.uum.edu.my/cancer

    21. Model of WEBCANDI

    22. Components of WEBCANDI The server component is divided into 3 parts: Knowledge base Working Memory Inference Engine

    23. Components of WEBCANDI Knowledge base Contains of the domain knowledge that was acquired from Dr. Zabedah Othman, our domain expert. We use interviewing technique to elicit the knowledge. The knowledge are represented in the form of production rules (IF/THEN format)

    24. Components of WEBCANDI Working Memory Contain the facts about a problem that are discovered during a consultation. The facts enter by the user will be stored into the working memory.

    25. Components of WEBCANDI Inference Engine Inference engine is the knowledge processor of the expert system It matches the facts contained in the working memory with the domain knowledge contained in the knowledge base, to draw conclusion about the current problem.

    26. Welcoming Screen

    27. Explanation Facility WEBCANDI provides 2 types of explanation facility: WHY it is asking a question HOW it reached some conclusion.

    28. Demonstrating WHY

    29. Demonstrating WHY

    30. Demonstrating HOW

    31. Demonstrating HOW

    32. One of the real benefits of web based expert systems is the ability to deliver real expertise to people who need it even though they are in remote locations WEBCANDI is a prime example of this application. WEBCANDI does not extend the state-of-the-art in AI, but it is the first expert system written using cold fusion mark up language (CFML), a web programming language. Rather, the system performed simple tasks since this is the first attempt to explore the practicality and applicability of Cold Fusion in developing intelligent web-based system. Conclusion

    33. Conclusion WEBCANDI has been tested and verified by Dr. Zabedah Othman by providing satisfactory advice and reasonably resemble the way she think. Although WEBCANDI is still in infancy stage of development, WEBCANDI does provide an early diagnosis of cancer for simple and limited cases. WEBCANDI recommend the user to seek professional advice if the user is positively diagnose with cancer

More Related