A Mobile-Based Fuzzy Expert System for Diagnosing COVID-19
Abiodun Muyideen Mustapha, Temitope Elizabeth Ogunbiyi, Deborah Aleburu, Christianah Yetunde Alonge · 2024
In 2020, COVID-19 was a great pandemic that affected the entire world. Millions of people across the world died and businesses were affected. Several individuals have had to spend very heavy medical bills or lose their lives as a result of late diagnosis. This is because the general diagnosis approach is based on medical practitioners' practical involvement. In this research, we present a mobile-based expert system for diagnosing COVID-19 using a Mamdani fuzzy symptom classifier that serve as an alternative to medical laboratory diagnosis. Three phases were involved in this development and they are: information elicitation and feature extraction (validation), knowledge base establishment, and information retrieval via the user interface. The mobile App was built using Flutter, Dart, and MySQL. The front end of the system, which is the mobile application, was built with Flutter. The system's front end was written in Dart, while the back end, which contained the COVID-19 data set, was written in MySQL. The identified inputs are the symptoms that were run through an inference engine and fuzzy inference rule to ascertain the severity of the disease in the patients. The result shows that the system was able to diagnose the disease through an interactive questions and responses from the user interface by the user. It becomes easy for medical doctors to recommend solutions to them. The evaluation performance of the system returned very high accuracy and was efficient in its diagnosis process. The result clearly indicates that the system is well accepted. Being a mobile-based system, accessing and engaging in its experience is faster.