AI-Integrated Probabilistic Neuro-Fuzzy TemporalFusionNet for Robotic IoMT Automation in Chronic Kidney Disease Detection and Prediction

Background Information: Chronic Kidney Disease (CKD) continues to be a major health burden across the world and hence early detection and constant monitoring of CKD is important for proper management. Previous diagnostic methods are usually insufficient due to the uncertainty of medical data, poor scalability, and cannot process in a real time way. The coupling of Artificial Intelligence (AI), Internet of Medical Things (IoMT), and advanced neuro-fuzzy systems provides a transformative targeted approach to CKD management. Moreover, this study introduces a strong framework for CKD detection and prediction using Probabilistic Neuro-Fuzzy Systems integrated with Artificial Intelligence (AI) along with Temporal Fusion Net (TFN). Goals include diagnosis in real-time, scalability, and interpratabilility via the robotic IoMT automation. The framework works through feature extraction from multi-modal IoMT data followed by processing of temporal information via TFN. While Neuro-fuzzy reasoning deals with uncertainty, robotic IoMT aids in the automation of monitoring and intervention. Performance measures can be accuracy, scalability, latency and interpretability. The framework produces an accuracy of 97.5% in prognostic prediction, a reliability of 96.8% in probabilistic risk prediction, and a real-time inference time of 0.85 s In terms of scalability and interpretability, it achieves the highest data fusion efficiency (92.7%) and a higher number of IOs per each identified relationship in finer-grained abstract model representations (9.6/10) for 1,500 IoMT devices compared with representative baselines among hybrids. This framework helps to manage CKD through a fast, explainable, and scalable solution which not only improves the patient care through better CI but also enables guided personalized healthcare.