Dermedys: Android-based dermatological diagnostic and analytic expert system

This paper describes a study to develop a ubiquitous expert system that aims to diagnose dermatological ailment present in a user. Diagnosing dermatological ailments has proven challenging due to the characteristic similarity in the symptoms of various diseases which usually require biopsies that are difficult to accurately read. This has brought about artificial intelligence researches into the automated means of accurately diagnosing such diseases. The algorithms, which have been designed although with nearly 98% of accuracies, are usually too heavy to work on pervasive systems, and also require histopathological inputs from the user. In this study, a system implementing a backward chaining algorithm using logical inferences, abduction, deduction, and induction was designed in such a way that it is lightweight enough to run on mobile phones. An android application, `Dermedys' with a knowledge base of five erythemato-squamous diseases was developed to implement the algorithm.