Fuzzy Logic Inference System for Managing Intensive Care Unit Resources Based on Knowledge Graph

With the rapid growth in the availability of digital health-related data, there is a great demand for the utilization of intelligent information systems within the healthcare sector. These systems can manage and manipulate this massive amount of health-related data and encourage different decision-making tasks. They can also provide various sustainable health services such as medical error reduction, diagnosis acceleration, and clinical services quality improvement. The intensive care unit (ICU) is one of the most important hospital units. However, there are limited rooms and resources in most hospitals. During times of seasonal diseases and pandemics, ICUs face high admission demand. In line with this increasing number of admissions, determining health risk levels has become an essential and imperative task. It creates a heightened demand for the implementation of an expert decision support system, enabling doctors to accurately and swiftly determine the risk level of patients. Therefore, this study proposes a fuzzy logic inference system built on domain-specific knowledge graphs, as a proof-of-concept, for tackling this healthcare-related issue. The system employs a combination of two sets of fuzzy input parameters to classify health risk levels of new admissions to hospitals. The proposed system implemented utilizes MATLAB Fuzzy Logic Toolbox via several experiments showing the validity of the proposed system.