Traditional healthcare systems face significant challenges, including delayed disease diagnosis, limited remote access, and severe data privacy vulnerabilities when centralizing patient records. To address these critical issues, this research proposes an advanced, comprehensive IoT-Driven Name-Based Patient Monitoring System integrated with Federated Learning (FL) and Artificial Intelligence (AI). The system utilizes an array of Internet of Things (IoT) sensors to continuously harvest real-time physiological data, including body temperature, oxygen saturation (SpO2), heart rate, and blood pressure. Furthermore, the diagnostic architecture extends beyond tabular data by incorporating Convolutional Neural Networks (CNN) to process complex unstructured inputs such as patient cough audio (via MFCC feature extraction) and X-ray imaging. To mitigate the substantial privacy risks inherent in traditional centralized machine learning, the proposed framework employs Federated Learning. This decentralized approach ensures that raw, sensitive patient data remains strictly on local hospital servers or edge devices, transmitting only cryptographic model weights (updates) to a central cloud server for Federated Averaging. Additionally, to safeguard the infrastructure against malicious network intrusions, a dedicated Distributed Denial of Service (DDoS) detection model

and Key-Based Authentication protocols are embedded within the network layer. Experimental evaluations demonstrate that the synergistic integration of ML algorithms (KNN, SVM, Random Forest) and Deep Learning models yields high predictive accuracy while maintaining strict data confidentiality. Ultimately, this highly scalable and secure ecosystem significantly reduces hospital visit costs, enables early disease detection, and fosters a smarter, decentralized digital healthcare environment.

Keywords— Internet of Things (IoT), Federated Learning, Artificial Intelligence, Disease Prediction, Name-Based Tracking, Cybersecurity, DDoS Detection, Healthcare.