Abstract:
Chronic kidney disease (CKD) is an abnormality of kidney function that is present for
xii
more than three months and causes damage to both kidneys and continues for a long time and know a day it is one of the major health problems throughout the world which cause for death. Chronic kidney disease includes five stages. In order to decrease the rate of death from chronic kidney disease there should be a fast and effective system for diagnosis, treatment and prevention especially, in developing countries like our country where there is a shortage of experts.
KBS is one of the mechanisms that can solve the problems that our country is facing right now. This mechanism can decrease the death from chronic kidney disease or failure by assisting and supporting the medical workers and patients in diagnosis, treatment and prevention of the chronic kidney disease for each stage.
The objective of this study is to design a knowledge-based system for CKD diseases. This KBS is composed of knowledge base, inference engine and user interface. This research mainly presented the development of a knowledge-based model that objects to provide the Patients/physicians with medical advice or support for diagnosis, treatment and prevention and basic knowledge on the five stages of chronic kidney disease. This knowledge-based model mainly takes a set of symptoms of the chronic kidney disease, GFR, ACR and also consider the age of the patient to diagnosis and identify the kidney disease with the stages and to order the different drugs as a treatment and also to take the prevention mechanisms.
In this thesis, we use the knowledge acquisition which is a knowledge-based approach to get facts using interviews from domain experts using purposive sampling technique from Felege Hiwot Referral Hospital and other sources, such as different web sites. The acquired knowledge from domain experts and document analysis is modeled using rule-based reasoning approach and represented by CommonKADS methodology and implemented using swi_prolog 8.0.3 tool.
The proposed knowledge-based system is assessed and evaluated by six medical and health experts using user acceptance testing by preparing nine questions and six domain experts using system performance testing by preparing 20 test cases. The developed knowledge-based model achieves 86% result of user acceptance and 90% result system
xiii
performance. The total performance of the proposed system is 88%. Developing in a better interface programming language, in an integration technique and using local language like Amharic are future work of the study.