DEVELOPING AN EXPERT SYSTEM FRAMEWORK FOR SUPPORTING DIAGNOSIS AND TREATMENT OF DYSPEPSIA AND GASTRIC CANCER

Abstract

Health care is one of the vital components of social services that have a direct relation to the growth and development of a country as well as to the welfare of the society. Ethiopian government has implemented different health policies to improve health care service, from that dyspepsia and gastric cancer is the most common internal human disease. Dyspepsia is a pain of the upper abdominal that has the symptoms of heartburn, nausea and upper abdominal fullness and this abdominal pain can cause gastrointestinal disease especially gastric cancer. Gastric cancer is the cell in stomach that begin to grow out of control and it can spread to other part of human body which result the death. Consequently, such type of disease requires timely diagnosis and treatment otherwise such type of disease can cause death and other chronic diseases. But still now in developing countries like Ethiopia there is poor the treatment option for dyspepsia and gastric cancer, to solve the scarcity of medical professionals, the concept of medical expert system can play an important role to support the disease diagnosis and treatments option. Therefore, the research study aims to develop an expert system framework for supporting diagnosis and treatment of dyspepsia and gastric cancer using local language (Amharic language), In order to achieve the general objective of the research and to answer the research questions knowledge engineering research design and qualitative research approach are used. This research study aims to develop an expert system for supporting diagnosis and treatment of dyspepsia and gastric cancer as the scope of the research. To develop this expert system, the domain knowledge was acquired using an interview from domain expert which are selected using purposive sampling techniques from Arba Minch General Hospital, and secondary data was acquired from a different source as explicit knowledge. After that the acquired knowledge is modeled using decision tree and rule-based knowledge representation is used to include the a numerous of rules into research study by using production rule. This expert system is developed by using backward chaining to infer the rule and SWI-prolog tool is used. Similarly, the acquired domain knowledge is stored in the knowledge base of an expert system and accessed through the user interface using local language for increasing end-user acceptance. Finally, the performance of an expert system prototype is evaluated using test case and user acceptance testing. Based on the evaluators result system scored 80% and 85.4% respectively. The total performance of the proposed expert system is 82.7%. This implies that the system has promising result. The researchers recommended to integrate data mining techniques with knowledge-based system for better performance of the system. Keywords: Diagnosis, Dyspepsia, Expert system, Framework, Gastric cancer, Treatment