Thesis Open Access

A Hybrid knowledge based clinical decision support system using Rule based and Case based reasoning for HIV patient who start ART medication: The case of Debre Birhan Referral Hospital

Yared, Bahru

Ethiopia has been providing free Antiretroviral Treatment (ART) since 2005 for HIV/AIDS patients. ART improves survival time and quality of life of HIV patients but ART treatment outcomes might be affected by several factors. However, factors affecting treatment outcomes are poorly understood in Ethiopia. There are different ways of managing ADR case of HIV/AIDS. Knowledge Based System (KBS) is the widely used one with rule-based reasoning or case-based reasoning. In this study, a combination of rule based and case based reasoning for managing ADR case of HIV/AIDS is proposed. To this end, knowledge is extracted using semi-structured interview from domain experts which are selected using purposive sampling technique from Debre Birhan Referral Hospital ART dep't. Relevant documents analysis method is also followed to capture explicit knowledge. Then, the acquired knowledge is modeled using decision tree that represents concepts and procedures involved in managing ADR case of HIV/AIDS and production rule is used to represent the domain knowledge. The method of combination used is a conditional combination model, which has a controller in between RBR and CBR. The knowledge-based system is developed using the integration of two programing languages with the interfacing library of JPL. A Prolog programming language with a SWI-Prolog tool is used to construct the rule base module, Java programming language with eclipse IDE tool is used to develop case base module, controller and GUI. In the combined system, it is the RBR that first treat the new query for recommending a solution. Otherwise, the query is automatically forwarded to the CBR system where the case retrieval module identifies the most related solution using case similarity measure. The combination of rule-based and case-based reasoning methods has shown a substantial improvement with regards to performance over the individual reasoning methods. The combined system scores 88.92% overall performance and achieves 87.5% accuracy with an average Precision and Recall of 87.5% and 91.67% respectively. The user acceptance testing also resulted 90.33% this is a very good acceptance. This shows the system has registered a promising result to come up with an applicable system. But, further exploration has to be done to refine the knowledge base and boost the advantages of combining RBR with CBR.

Files (2.9 MB)
Name Size
2.9 MB Download
All versions This version
Views 253253
Downloads 100100
Data volume 294.4 MB294.4 MB
Unique views 150150
Unique downloads 7979


Cite as