Diagnosing Support Expert System for Symptomatic Fever Onset of Kawasaki Disease

Distinct from the classical fever temperature range, healthcare professionals might deal with a sort of febrile illnesses that encompasses multi-characteristic analysis, such as Kawasaki disease. This paper aims at discussing the fever symptom from a multidimensional perspective to support healthcare professionals detect illnesses onset of Kawasaki disease by using an expert system. The set of diagnostic linguistic variables comprising the intensity of the temperature, time of onset, knowledge of origin of infection, and efficacy of antipyretic treatment, as preconized by the American Heart Association criterion, are aggregated by employing a set of IF-THEN rules designed according to the experience of healthcare professionals. This paper innovates in proposing an alternative model based on artificial intelligence approach that enables analyzing a set of diagnosing variables that compose the fever onset of Kawasaki disease. The resulting expert inference system provides qualitative assessments, through staging, from the admission of values at intermediate levels. The assessment of risk and severity for the mandatory symptom fever in Kawasaki disease using the expert system is an improved alternative to the American Heart Association criterion. The proposed diagnosing expert system for Kawasaki disease's symptomatic fever can be used as a tool for supporting healthcare professionals to the disease early detection in order to avoid worsen onset of such a critical clinical condition.