Application of Fuzzy Inference Systems in Predicting Patient Outcomes in ICU Settings

This paper presents a fuzzy logic-based approach for predicting the risk level of patients in Intensive Care Unit (ICU) environments using key physiological indicators: Heart Rate, Glasgow Coma Scale (GCS), and Oxygen Saturation (SpO?). A Fuzzy Inference System (FIS) was developed incorporating expert-defined membership functions and a comprehensive rule base consisting of 27 IF?THEN rules that represent combinations of clinical inputs and their impact on ICU risk. Defuzzification using the Mean of Maximum (MOM) method converts the aggregated fuzzy output into a single crisp ICU risk score. Three-dimensional surface plots illustrate the dynamic interaction between input parameters and the resulting ICU risk levels. The results show that combinations such as high heart rate, low SpO?, and low GCS are associated with elevated ICU risk, while normal physiological ranges correspond to low risk. The proposed model demonstrates the potential of fuzzy logic systems to support real-time, interpretable decision-making in critical care settings.