OFNN-UNI: Enhanced Optimized Fuzzy Neural Networks Based on Unineurons for Advanced Sepsis Classification

Merging machine learning with clinical datasets has facilitated advancements in managing sepsis, a medically critical condition. This study introduces the Enhanced Fuzzy Neural Network Methodology with UniNeurons (OFNN-UNI), distinguished by interpretability and efficiency Achieving an accuracy of approximately 93%, the OFNN-UNI’s performance is on par with contemporary state-of-the-art models, emphasizing the crucial role of accurate sepsis detection in medical intervention. The model, refined by Stochastic Gradient Descent for more precise rule consequent definition, applies fuzzy logic to navigate data ambiguities. By analyzing variables such as patient age, gender, and sepsis history, it provides insights through the definition of fuzzy rules and data trends, marking a step forward in predictive analytics for sepsis.