Fuzzy Neural Networks for Real-Time Decision Support in Healthcare

For AI-based healthcare systems to be more interpretable and accurate, clinical decision-making complexity is increasingly escalating and hence becomes a significant research issue to examine. The hybrid fuzzy neural network that combines fuzzy logic, for clinical uncertainty management, with deep learning, for noise-resilient pattern discovery. Adaptive patient profiling is derived from a self-adjusting fuzzy-clustering algorithm, and contrastive self-supervised learning learns features from incomplete medical records; then a rule-augmented fuzzy inference layer outputs explainable reasoning. Executed through Edge-AI and federated learning, the model outputs real-time predictions without infringing on patient privacy. Measured on the Comprehensive Disease Symptom and Patient Profile Dataset, the HFNN achieves 94.6 % accuracy and 0.97 AUC–ROC, beating state-of-the-art deep nets and single-expert systems. In combining accuracy, adaptability, and explainability, the paradigm brings healthcare closer to reliable, time-series-ready clinical decision support. Traditional deep learning models achieve high diagnostic performance but operate as black boxes, hindering trust in medical applications.