Expert system for early diagnosis of eye diseases infecting the Malaysian population

Abstract

Describes a knowledge-based system employing certain expert system rules to detect different kinds of eye diseases found in Malaysia. The types of eye diseases that can he detected with this system are allergic or infectious conjunctivitis, secondary and senile cataracts, open-angle glaucoma and acute glaucoma, keratitis and "dry eyes" syndrome. These are the most frequent eye diseases infecting the Malaysian population. The project was designed and programmed via the object-oriented expert system shell EXSYS. The expert rules were developed based on the symptoms of each type of eye disease, and they were presented using tree-graph forward-chaining with a depth-first search method. In order to enhance user interaction with the system, graphical user interfaces were employed. Previously, several similar systems have been presented, but they were limited to detecting a single disease and also required an expert medical officer to operate them. The expert system described in this paper is able to detect and give early diagnosis of five types of eye diseases, including senile, secondary, open-angle, acute, allergic and infectious diseases.