Research On Computer Inference Systems With Mechanisms Of Computerised Malaria Diagnosis In The Tropics

Medical Diagnosis is the clinical discovery or exploration of what kind of illness an individual is suffering from basing on the exhibited signs and symptoms. Diagnosis of Malaria diagnosis, therefore, can be carried out in many ways. One of the ways is by visiting a health facility to have clinical tests to be carried out so as to confirm the infection type affecting the patient. In many places in Africa, endless people die annually due to lack of medical facilities to carry out the clinical tests. Malaria being an infectious disease and number one killer disease in Africa, it has caused intolerable challenges to the already strained politicians.As scientists, we discuss ways of how to try and solve this problem in a feasible and more so scientific approach. Considering the fact that the health sector in Africa, Uganda in particular is understaffed, the way discussed here is basically to relieve the present or existing efforts to combat the problem. We propose a mechanism of involving knowledgeable and computer literate people to help in the medical/health service delivery. This is by diagnosing what one is suffering from given little historical background or information about their health problems. We make use of the tools and techniques of inferencing.Inferencing is the intelligent reasoning on available data help make a decisive conclusion about a certain subject. Meaning Medical Diagnostic Inferencing is the use of an intelligent reasoner in helping identifying what one is suffering from. The several Inference methods known include the use of Inference Rules or using the Bayesian Approach. The latter is a more deterministic approach while the latter is more probabilistic. In this paper, we capitalize on using the latter mechanism by constructing a Knowledge Base in Protege for carrying out Malaria Diagnosis. We designed a diagnosis and a treatment algorithm, specifically for Malaria. As a result, this thesis presents the results included here by testing the work using the Belief Networks Builder and comparing our results with PROMEDAS.