A Belief Rule-Based Expert System to Diagnose Influenza

Influenza is a viral disease that usually affects the nose, throat, bronchi, and seldom lungs. This disease spreads as seasonal epidemics around the world, with an annual attack rate of estimated at 5%–10% in adults and 20%–30% in children. Thus, influenza is regarded as one of the critical health hazards of the world. Early diagnosis (consisting of determination of signs and symptoms) of this disease can lessen its severity significantly. Examples of signs and symptoms of this disease consist of cough, fever, headache, bireme, nasal congestion, nasal polyps and sinusitis. These signs and symptoms cannot be measured with near-100% certainty due to varying degrees of uncertainties such as vagueness, imprecision, randomness, ignorance, and incompleteness. Consequently, traditional diagnosis, carried out by a physician, is unable to deliver desired accuracy. Hence, this paper presents the design, development and application of an expert system to diagnose influenza under uncertainty. The recently developed generic belief rule-based inference methodology by using the evidential reasoning (RIMER) approach is employed to develop this expert system, termed as Belief Rule Based Expert System (BRBES). The RIMER approach can handle different types of uncertainties, both in knowledge representation, and in inference procedures. The knowledge-base of this system was constructed by using records of the real patient data along with in consultation with the Influenza specialists of Bangladesh. Practical case studies were used to validate the BRBES. The system generated results are effective and reliable than from manual system in terms of accuracy.