Skip navigation

Please use this identifier to cite or link to this item:
Title: Knowledge Based System for Maternity Care, Treatment and Prevention
???metadata.dc.contributor.*???: Dr. Dereje Teferi
Kebede, Bisile
Issue Date: Jan-2017
Publisher: Addis Ababa University
Abstract: In developing countries like Ethiopia, fulfilling every aspect of health issues is an immense challenge. These challenges are confronted in different sectors by various methods. The stakeholders include governments, non-governmental organizations and private sectors. Among the efforts of United Nations organization to fulfill this goal includes, setting international plans such Millennium development Goal, Supporting countries in financial, human resource, training and infrastructure. Despite of all these efforts, maternal mortality ratio in sub-Saharan Africa is still in its highest rate in the world. In Ethiopia, MMR (Maternal Mortality Ratio) in 2013 is 420 deaths per 100,000 live births. United Nations set to reduce this number to 350 deaths per 100,000 live births in 2015. Major challenge for health service in Ethiopia is shortage of skilled manpower in the health sector. Quality of Maternity care is compromised due to services provided by less skilled personnel. As part of the stakeholders in this regard, information technologies play a significant role. The objective of this system is to develop a knowledge base system that assists Health workers in decision making process related to maternal cases. The limitations of human power are overcome by the diligence of computer systems like this. This system can be used by low level to high skilled health professionals. Hence, consistency among maternity care decisions is achieved. The quality of service to maternal women is also improved. To achieve the expectations of this system, the methodologies followed in this system are knowledge acquision using Interview is used as one of primary knowledge source which we can gather the needed knowledge from knowledge experts such as Doctors, Gynecologists, and Midwifery Nurses face-toface. Domain experts were selected by using purposive sampling techniques. Semi-structured interviews can be used to gather qualitative information. For the research will use decision tree modeling in formal language which is suitable for rule based system. From decision tree the prototype is developed using SWI-prolog, which uses rule based („if - then‟) rules. After the development of knowledge based system verification and validation tests are conducted. The verification test of knowledge based system is tested in two categories, code standard by ICT staff and user acceptance by domain experts. According to ICT staffs 80% of the code is up to the standard and 82% of the users are satisfied with system. In addition, the performance of the system is evaluated by using validation techniques with twenty test cases, which five cases are assigned for each maternal visit. The results of the validation test case indicate that 90% of cases are similarly treated. The main challenges faced during the studies are medical knowledge is difficult to understand with in short time, the schedule of the experts are crowded and others are discussed. Finally, some recommendations are forwarded.
Description: A Thesis is Submitted to the school of information science Addis Ababa University in Partial Fulfillment of the Requirements for the Degree of Master of Science in Information Science.
Appears in Collections:Thesis - Information Science

Files in This Item:
File Description SizeFormat 
BISILE KEBEDE.pdf1.18 MBAdobe PDFView/Open
Show full item record

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.