ARTIFICIAL INTELLIGENCE-BASED EXPERT SYSTEM FOR DIPHTHERIA DIAGNOSIS

Conventional diagnostic techniques frequently depend on the observational conclusion of lab results and clinical indicators, which can cause delays. Consequently, there is a need for a fast, precise and cost-effective technology to diagnose infectious diseases and thus reduce morbidity and mortality in the under-developed and developing countries. Although numerous research in the medical domain have been conducted by different researchers utilizing diverse diagnosis approach, In this study, A V-P Expert System shell was utilized in the creation of a diphtheria diagnosis system (DDS). This is a rulebased system that employs forward-chaining approach for diagnosis. It consists of three modules, the User Interface to facilitate user interaction, allowing input of queries and displaying result, the Inference Engine to process queries and applies rules to derive conclusions and lastly, the Knowledge Base to store facts, rules, and relationships about the domain. The knowledge base was created by compiling accurate knowledge from medical experts in diphtheria. The system offers a simple, interactive user interface, where diagnosis of patient is achieved based on microbiological and clinical examinations. Preliminary testing shows that the system provides consistent and reliable diagnostic support, making it beneficial for remote areas with limited access to medical experts. However, its accuracy depends on the completeness of the knowledge base, and it may be less effective in cases with atypical symptoms. Future improvements could include expanding the knowledge base and integrating adaptive learning techniques. This automated approach enhances diphtheria detection in underdeveloped regions, improving diagnosis and treatment outcomes.