A Fuzzy Logic-Based Framework for Accurate Detection of Infectious Diseases

The proposed framework is designed to enhance the early detection and diagnosis of diseases, aiming to improve treatment efficiency and boost recovery rates. It operates by processing data collected from wearable devices through a fuzzy inference system (FIS) consisting of key components: fuzzifier, Knowledge Discovery in Databases (KDD), a predictor, and defuzzifier. Fuzzifier renovates raw input data from wearable devices into fuzzy values suitable for use within the fuzzy inference system. KDD analyzes the fuzzified data to uncover important patterns and insights necessary for training the prediction model. The Predictor utilizes both the fuzzified data and insights from KDD to train a model capable of detecting diseases based on the processed data. Defuzzifier then translates the fuzzy output from the prediction model into clear, actionable values. The research employs the Weka tool to analyze a dataset related to monkeypox. Using the J48 algorithm, a fuzzy decision tree is constructed, and eight fuzzy rules are established based on the C4.5 algorithm. With approximately 25,600 instances in the dataset, 16,682 are correctly classified (66.728%), while 8,318 are incorrectly classified. This fuzzy logic-based approach aims to expedite disease detection processes, thereby reducing the need for hospitalization. Furthermore, it seeks to enhance the interaction between healthcare providers and patients, ultimately improving overall healthcare delivery.