Development of an Expert system for the diagnosis of pneumonia in children

Description

Abstract This research work focuses on an expert system for the diagnosis of pneumonia in children. The aim of this project was achieved using input symptoms associated with pneumonia are fever, loss of appetite, fatigue, cough, chills, haemoptysis, fast breathe and chest pain to model its input with the pneumonia level as the output. The triangular membership function was employed in this project to determine the degree of membership of both the input and output in the fuzzification process. The Mandani max-min inference engine was used for the logical reasoning using the fuzzy output back to crisps out in the defuzzification process. Furthermore, the application was developed using Java programming language due to its flexibility, robustness and its ability to handle fuzzy logic system effectively. The final result of this project will enhance medical practitioners in performing fast and reliable diagnosis of pneumonia.