Adaptive deep neuro-fuzzy systems for real-time medical diagnosis using wearable IoT devices

The rapid proliferation of wearable Internet of Things (IoT) devices and advancements in artificial intelligence (AI) have reshaped healthcare diagnostics by enabling non-invasive and real-time monitoring of physiological parameters. This article proposes an adaptive deep neuro-fuzzy inference system (ADNFIS) integrated with wearable IoT technologies for intelligent real-time medical diagnosis. The proposed hybrid model leverages the reasoning power of fuzzy logic, the learning capabilities of neural networks, and the feature extraction efficiency of deep learning to handle nonlinear, imprecise, and dynamic medical data. Health-related signals, such as heart rate, oxygen saturation, electrocardiography (ECG), and electroencephalography (EEG), measured by wearable sensors, are continuously collected, pre-processed, and transmitted through a secure IoT framework to edge or cloud computing for diagnosis. The neuro-fuzzy architecture adapts over time to patient-specific patterns, enhancing diagnostic accuracy and reducing false positives in real-time scenarios. The system was evaluated using publicly available wearable health datasets, demonstrating superior performance compared to traditional models in terms of accuracy, response time, and robustness to noise. Additionally, the model’s high transparency and interpretability offer significant benefits for clinical decision-making. The integration of wearable sensing, adaptive learning, and explainable AI in the proposed framework represents a step toward personalized and continuous healthcare monitoring. Future work will extend the study to multi-modal inputs and incorporate edge AI optimizations to minimize latency and enhance privacy.